1. Introduction
[1.1] Being the products of convention rather than nature, all cultural artifacts—literary, linguistic, or otherwise—pull in and push against the ideas, values, tropes, and words embodied in and projected onto what we might call their canonical frames (note 1). This push-pull action is a diverse dialectic (incorporating varying levels of intentionality) of adherence to claimed or perceived values and forms and the transformation of such values and forms, a dialectic of affirming and rejecting continuity, repetition and homage, appropriation and transvaluation. In this general sense, we can describe many, perhaps all, literary and cultural artifacts as fan works and their producers and receivers as fans. We need only replace the word fans with the words devotees, cognoscenti, or interlocutors, or with the term "members of interpretive communities," to see that these claims are less provocative than they first appear (note 2). Virgil, Dante, and Joyce are as much fans of Homer and the epic tradition and as much textual poachers (de Certeau [1975] 2000; Jenkins 1992) as self-professed 16-year-old fan Avaron, who posts a fan fiction response to the Iliad. "It was never about the war," her story's epigraph reads, with the disclaimer, "I do not own the Iliad (though I wish I wrote [sic] like Homer)" (https://www.fanfiction.net/s/11201974/1/glory-gore). An identical dialectic is in effect, whether we are dealing with ancient or modern materials or media, with classical poetry and the canonical foundations of Western literature, or with recent fan fiction.
[1.2] Such universal observations, if they are not truisms, threaten us with the bleak prospect of a uniform, ahistorical world and with little critical traction. If notions of modernity, of fan fiction as a contemporary phenomenon, or of antiquity or classical literature are to have any meaning, they must be historicized, defined historically not merely in deictic terms but conceptually, in a manner that exposes difference and that relies, at least to some degree, on exclusive distinctions. The risk here is that by successfully marking off discreet historical or conceptual domains, we will have essentialized historical moments, sealed them off from each other, or created mere historical marionettes.
[1.3] By way of a response, I want to try to walk something of a middle road that on the one hand will highlight historical difference, allowing us to regain at least some critical-historical traction, but that on the other hand will expose meaningful isomorphism. To do so, I shall have to move away from the idea of canon as object or work and toward the notion of canon as a socially embedded practice of containment of surplus, in which the object itself has more flexible attributes. After a brief methodological comment and a short section on the historicity of the term canon, I shall take a closer look at an important recent discussion of canonicity in relation to one of the earliest and therefore potentially most distant and different domains of production and reception in the literary tradition of the West: the early Greek oral and orally-derived epic. By borrowing but also critically (and necessarily) inverting some of this discussion's perspectives, I want to reconsider how the fundamental push-pull dialectics of canon operate in such an early, radically different context and how these dialectics relate to the production and proliferation of discourse, to acts of preservation, and to acts of deletion and censorship. I also want to explore the relation of such specific dialectics to works and practices in other environments. As we shall see, under specific conditions, the idea of textual poaching (de Certeau [1975] 2000; Jenkins 1992) and the responsive positions underlying such fan fiction terms as "alternative universe," "original story," and "original character" are not quite so far from the classical beginnings of Western literature. Yet precisely such proximity can also provide us with a useful position from which to observe other, more complex, and more specific historical variance.
2. Categories of difference and historical compounds
[2.1] First, I offer a brief methodological comment. In formal chronological, medial, geographic, linguistic, and other terms, archaic Greece, our case in point, and its cultural outputs are distant from late modern digital cultures. Such terms as archaic, modern preliterate, digital, Ionian, and global necessarily underpin our historical understanding, but they are rarely sufficient in and of themselves. Often the closer we inspect them, the more problematic they become. For this reason, in what follows I focus less on essential attributes and distinctions and instead on the fragile facticity of historical compounds (note 3). The differences and similarities between the domain of archaic Greek epic and 21st-century online or other narratives have less to do with medial attributes (and, for example, with the relations between an oral performance, a stone slab, a roll of papyrus, a computer monitor, a Web site, or an electronic list), less to do with formal dating, language, or geography, and more to do with situated matrices and conditions of production and reception, with contextualized constructs about how poets, stone carvers, writers, readers, observers, passers-by, and online surfers do things with a text and with its canonical frames. Such frames are themselves framed and are thus themselves fragile historical objects, but precisely for this reason, I would suggest, they are the foundation of historical comparisons.
3. The historicity of canon
[3.1] Let me begin with this last observation. Fan fiction and also acafans, for example, sometimes take a practical view of canon (note 4). It's "the story as told by the original author," as Jamison (2013, 26), suggests; or as the online Fanspeak Dictionary puts it, yet simpler language, in the entry for the term canon, "Megabyte's real name being Marmaduke is canon because it expressly says in Origin Story that it is" (http://expressions.populli.net/dictionary.html, referring to The Tomorrow People [1992–95]). Students of classical antiquity often take an equally practical view of the canon as a simple list of undisputed masterworks, not least the works of Homer, the "founding hero (heros ktistes) of Western Literature" (Curtius 1973, 16).
[3.2] Nevertheless, in a more reflective frame of mind, many scholars have long been challenging the idea that, as Herbert Marcuse (1978) puts it, "throughout the long history of art, and in spite of changes of taste, there is a standard which remains constant" (Intro p. x) (cf. Szafraniech 2010). Canons and fanons evolve as the work of fans (in the sense of cognoscenti, as above) gains fans or supporters and respondents of its own and is inducted or subsumed into the canon while other work may lose fans and drop out of the canon (Maybin and Mercer 1996; Thomas 2007).
[3.3] More significantly, as scholars in disciplines as different as digital media, the study of modernity, and classics have argued, not only canonical tokens but also the very idea of canon are dependent on specific, historically contingent conditions of thought. Philosopher and music historian Lydia Goehr (1992), for example, discussing canon in music history in an argument with wider implications, has stressed the specific link of canon to emergent modernity. Around 1800, she says, the very idea of music underwent an essential ontological transformation. Classical music as we know it only came to be with the rise of Romantic aesthetics and German Idealism. These contingent frameworks of consciousness conceptualized repetition, especially in relation to performance, and hence reconfigured the notion of the musical work. "Johann Sebastian Bach," says Goehr, "did not intend to write musical works" (8) (cf. Kerman 1983). It is only since this era, Goehr suggests, that music begins to coalesce around the idea of a body of musical compositions and a canon—a valued and authoritative set of works at the heart of a culture.
[3.4] Such historical positioning chimes well with statements made about the classical tradition on the one hand and digital works and cyberspace on the other. Looking to the past, John Guillory (1983), for example, argues that canon is a term that "displaces the expressly honorific term 'classic' precisely in order to isolate the 'classics' as the object of critique" (6) (note 5). Some years earlier, the great historian of classical scholarship Rudolf Pfeiffer (1968) pointed out that the use of the word canon in the sense of an exemplary list of (secular) authors and works was not ancient but "a modern catachresis that originated in the eighteenth century" (note 6) introduced by Dutch classicist David Ruhnken (1723–98). Ruhnken, Emmanuel Kant's (1724–1804) fellow student at Königsberg, anachronistically projected the idea of canon onto Hellenistic scholarship (note 7). His thought is inherently modern and is inextricable from modern (rather than ancient) notions of repetition and performance and from modern concepts of public teaching curricula (prescribing texts and subjects of study) and pedagogical cultures of internalized responsibility, rational idealism, and the modern self (Ruhnken 1768; Pfeiffer 1968; Gorak 1993; Hägg 2010).
[3.5] From a quite different perspective, a complementary argument about the historical fragility of the idea of canon emerges, for example, in comments about the digital humanities. Here fundamental notions of repetition and performance, and thus again of the ontology of the literary work of art, seem to have changed. As one traditional student of the book, Roger Chartier (1994), notes, "The library of the future is inscribed where all texts are assembled, and read on a screen…an opposition long held to be insurmountable between the closed world of any finite collection, no matter what its size, and the infinite universe of all texts ever written is thus theoretically annihilated" (89).
[3.6] Some 20-odd years after the publication of the words above, their future and its consequences for the canon and for the "closed world of any finite collection" seem to be already here. Not surprisingly, Astrid Ensslin (2007) acknowledges the need for an alternative type of canon (cf. Bolter 2001; Thomas 2007). One can productively set such terms as "displacement," "catachresis," or the "theoretical annihilation of long-held oppositions" side by side. Combined, they suggest that whether we look at the question from the perspective of the classical past, modernity, or the digital future, the idea of canon is both mobile and historically constrained. If we are to retain it as a useful comparative term, we need to strive toward an understanding that affects comparison within contingency.
4. A practice-based perspective on canon
[4.1] Glenn W. Most (1990), reflecting on the question of canon in early Greek epic, especially in relation to early Greek poetry and orally-derived poetry, offered a reformulation of the idea of canon—not as a lists of works that are sanctioned as the best but, from a functional point of view, as "the device that permits us to distinguish between the books we can admit without shame that we have not read and the books that we feel so ashamed about not having read that we usually feel obliged to pretend that we have" (37) (note 8). This idea takes the focus of the term canon away from any set of outside objects, which are always liable to change, and thus in part also away from ontological assumptions about what constitutes a work, which may also change. Most's formulation places the emphasis on what might be described as a phenomenology of the process, one that (though Most himself does not openly say so) underpins a sense of social belonging. This perspective helps us problematize and perhaps better understand not only ancient epic but also contemporary fan fiction. First, however, let us take a closer look at some of the details of Most's suggestion.
[4.2] Underlying Most's (1990) idea is the simple notion of a surplus of words and thus of the need for canonical regulatory apparatus. According to Most—and this is where his argument begins to gain some specific historical texture—what produces this surplus and thus the practice of canon is the medium of writing. A book, roll of papyrus, or wax tablet is a material object that holds words but is independent of the performer or the fleeting moment of performance. Writing allows words to accumulate. Thus, in literate cultures, Most argues, "canonization aims, not at the annihilation of the texts it designates as non-canonical, but only at positioning them lower on a scale of priority, at rendering their use less urgent" (44). I shall return to this idea later because it bears clear relevance for an appreciation of a wide range of canonical contexts (note 9).
[4.3] Here, however, Most (1990) makes a crucial further distinction: one between writing and orality. Oral cultures, he suggests, operate somewhat differently from ones that make use of writing: "An oral culture does indeed know some forms of relative ranking within hierarchies of canons…but within each genre the drastic consequences of relative unpopularity in an ideal oral environment lead inevitably to the irreversible displacement of less desired texts by more desired ones" (44, my emphasis). According to this argument, the medial/performative element of oral cultures forecloses the repetition of unpopular or less valued texts. We can sharpen this point by saying that according to Most, orality restricts the push-pull dialectic of the work of art to the synchronic moment of performance but largely curtails diachronic interaction.
[4.4] To prove his point, Most (1990) offers the example of a poem called the Nostoi, or Returns of the Heroes, from the Epic Cycle. The Cycle is a series of poems that supplement the stories of the Iliad and Odyssey and may well incorporate earlier traditions of oral epic song. Little of these poems survives—only some plot outlines and a handful of actual verses (note 10). Most argues that the oral process of performative annihilation in these written remains is responsible for the poems' scanty remnants: "Since they could not bear the competition with Homer, they were condemned to virtual extinction." Thus, Most concludes, "In an ideal oral society, where production and reception are indissolubly linked, canonization inevitably becomes censorship" (44). In oral cultures such as those of archaic Greece, in the presence of such medial censorship, neither canonical prescriptions nor the concept of canon as they later appear are thus necessary or indeed possible (note 11).
5. Losing the canon to history
[5.1] Defining the canon in these terms is an attractive means of gaining historical traction. We would seem to have a clear distinction between the early preliterate foundations of Western traditions on the one hand (which, strictly speaking, have no canon, or where canonization simply means censorship) and many of the canonical texts and mechanisms of later, literate Western culture, especially post-Enlightenment cultures, on the other (Guillory 1983, 1993; Goehr 1992; Gorak 1993; Kermode 2004). If Most (1990) is right, the embedded medial censorship of archaic orality is likely to have affected not simply epic poetry but also a much wider range of oral mythological, historical, ethnographic, and, in a broader sense, political, legal, and ethical discourses (note 12). Furthermore, we can perhaps also use his general method and assumptions, albeit in quite different ways, to separate the canonical practice of post-Enlightenment and print-based modern Western cultures from the practice of digital cultures in late modernity and perhaps also from the practice of fan fiction.
[5.2] Strictly speaking, fannish use applies the term canon to the source contents of franchised work—television series like Xena: Warrior Princess (1995–2001), the original Star Trek (1966–69), or the novels of Jackie Collins—which are heavily protected by intellectual property law and the ruthless commercial practice of corporate multinationals. Some mobility exists, of course. In a small number of cases, individual fan writers are co-opted into the legal/commercial enterprise of the franchise. But when fans are thus embraced, they become—precisely, formally—part of the canon (note 13). Fan fiction in its proper sense is not allowed into the domain propre of its source material. Here, it would appear—as in the case of ideal orality but for completely different, historically contingent, and specifically modern reasons that are inconceivable in antiquity—that canon draws dangerously close to the practice of censorship. From this perspective, fannish use of the term canon thus seems like a modern catachresis.
[5.3] Fandom's own domain propre is currently not television or books in print (pace fanzines) but the digital world online. Digital media, electronic bulletin boards, discussion lists, blogs, posts, and indeed fan fiction Web sites create what appears to be a surplus of material on a scale exponentially greater than ever before, a surplus that is increasing rapidly—arguably to such a degree that it affects not a quantitative but rather fundamental ontological or phenomenological changes to what constitutes the work. As Chartier (1994) says, "An opposition long held to be insurmountable between the closed world of any finite collection…and the infinite universe of all texts ever written is thus theoretically annihilated" (89). Inside this environment, despite frequent and explicit use of the term canon and despite external constraints, it is not at all clear what might constitute, for example, a general canonical character name or story line that, to use Most's (1990) words, we would "feel so ashamed about not having read" that we might "feel obliged to pretend that we have." We may know the names of Achilles and Marmaduke, but are these names more than ciphers of placeholders inside fan fiction? Would Avaron, whose work is "mostly just an aimless muse dump" (https://www.fanfiction.net/u/4683839/avaron-x), who posts online "after reading the Iliad," who wishes he or she "wrote like Homer" (https://www.fanfiction.net/s/11201974/1/glory-gore), admit or not admit that he or she has not read the Iliad, or has not read it fully or as closely as one might or as one should, if it is indeed canon? Amid the vast diversity of the digital world, does membership in a particular fan group, heavily mediated by aliases, depend on such acts of acknowledgment? Would Avaron feel ashamed—or indeed have any concern about never having watched—The Tomorrow People, or about not knowing who Marmaduke Damon or his synonym, Megabyte, is? Some judgmental views are clearly made in fandom. However, the range and quantity of the materials involved, the number of options open to readers and respondents, and, paradoxically, the number of local online communities involved are so great that the idea of an operative or indeed consistent canonical "scale of priority" seems nostalgic, if not downright impossible—or at least so if we follow the line advocated by critics like Chartier (1994).
[5.4] Have we, then, ended up with a narrow historical definition of the term canon, which, precisely for this reason, excludes both the beginning of our historical timeline and its most recent and future points from the realm of canonicity?
6. Regaining the canon through history
[6.1] I suggest that we can rearrange some of the useful aspects of Most's (1990) argument in a manner that allows us to acknowledge the difference of oral epic and fan fiction and at the same time to affirm and better understand their forms of canonicity. Let me first suggest how this might be conceptualized. I will then offer one or two concrete examples.
[6.2] Already Most's (1990) own argument hints at a more complex position. Consider first the question of early orality. There is no doubt that in oral settings less popular texts may be displaced by more desired ones. Nevertheless, already in antiquity, even the few remaining fragments of the Epic Cycle were presented as fragments of a larger text (most of the evidence we have today comprises plot summaries). Furthermore, even such scanty fragments were ascribed (rightly or wrongly) to named individual poets—Agias of Troezen (8th century BCE?) in the case of the Returns (Burgess 2001). Thus, at least some practical evidence of the canonical periphery clearly did survive beyond orality.
[6.3] More significantly, the principle of displacement that is the basis of the argument for de facto censorship and thus for the absence of canon in oral cultures points to the exact opposite conclusion. This has important implications for the question of canonicity in early oral cultures and by analogy for our understanding of canon in digital cultures. If we assume that in an oral environment it is possible to displace a narrative simply by not performing it, then by the very act of nonperformance, we can just as easily displace an officially sanctioned version as any other. A sanctioned text is only safe at one single moment, dependent on its immediate performance and the preference of the local community at that moment. In and of itself, this does not provide any text, sanctioned or otherwise, with the assurance of future performance. Thus, the potential for open-ended surplus is built into the idea of orality (note 14). As Michel de Certeau says, "Orality insinuates itself, like one of those threads of which it is composed, into the network—an endless tapestry—of a scriptural economy" (1984, 132). Out of this surplus—contrary to Most's (1990) conclusion but paradoxically according to his own principles—emerges, precisely, the idea of canonicity.
[6.4] Even the most distinct (almost scriptural) result of ancient oral censorship as Most (1990) defines it—Homer's poetry—openly states this principle of surplus and its pluripotent nature. In the first book of Homer's Odyssey, for example, we find the Ithacan singer Phemius rising to perform his song about the "Achaians' sad return [nostos] from Troy" (1.326–27). Likewise, when a tearful Penelope, wife of the absent Odysseus, protests that this song distresses her and urges Phemius to stop, her son, Telemachus, censures her for begrudging the singer to entertain the audience in "whatever way his [Phemius'] mind is spurred" (1.347). "Men," says Telemachus (who represents the future of the heroic past), "always praise the song that is newest" (1.351–52). He then immediately stresses the diversity of possible fates and narratives: "Odysseus was not the only one to lose his day of return [nostimon, from nostos, "return"] in Troy; many other men also perished." Here is precisely the missing tradition of the Nostoi, or Returns of the Heroes from the Epic Cycle.
[6.5] Already in Homer's written but unquestionably orally-derived poetry we see the dynamic diversity of orality, which is a precondition of the need to regulate and canonize. In oral contexts, almost by definition, we can introduce new, previously unsanctioned discourse more easily than we might in contexts that preserve official records and material texts (Foley 1988, 1991, 1999; Zumthor 1990). Pace Most (1990), in oral contexts, at the point of performance, there is thus always a surplus of possible texts competing for the audience's ear. What rounds off the argument for surplus is the fact that in oral environments the synchronic moment of performance, the moment of composition in performance, is identical to the moment of tradition. This moment is as indiscriminately fragile as it is superabundant, both because of the generative potential of oral performance and because phenomenologically, tradition is surplus; it always contains more than is said or can be said at any one moment of performance. Under such conditions, orality's excess must be contained by the notion, real or perceived, of a canonical frame. In the case of heroic epic, this frame is the work of the poet to whom we and the ancients give the name Homer (note 15).
[6.6] The idea of oral abundance also offers us an important point of contact with the surplus of the modern digital world and the world of fan fiction canons and fanons (note 16). The open-ended variety, accessibility, and ease of dissemination in the digital world in one sense suggests a breakdown of the opposition between finite corpora and anything "ever written" and thus the collapse of canonicity. Against the background of strong legal and commercial constraints from the franchise, fan fiction nevertheless claims its freedom to invent its own original material. Yet such rebellious declarations are fundamentally embedded in the recognition, inherent and self-generated, not imposed, of canonical frames. The name fandom (the fannish kingdom) suggests not an unstructured, lumpy universe but rather a structured regime. Even more revealing is the term fanon, which is unimposed from the outside; fanon is self-proclaimed and explicitly derived from canon. Fanon works in synergy with canon. It suggests a hybrid, even transformative, practice of containment. Here is one definition of the term fanon, also drawn from the Fanspeak Dictionary: "Things that are not strictly canon but do not contradict it and are widely accepted by most fans. For instance if most fans just accept that Megabyte's middle name is Archibald, even though it is not expressly canon, it becomes fanon" (http://expressions.populli.net/dictionary.html).
[6.7] In the environment of online digital texts, as in oral environments but for wholly different, historically specific reasons, in completely different medial and other contexts, we find a proliferation of options. These are numerous and diverse, and are located in an essential surplus of repositories and sites. Many such sites are self-selecting and specific to groups, subgroups, and smaller online communities (Jenkins 1992; Hellekson and Busse 2014). All this generates a surplus of material that authors can produce and publish easily and rapidly and that, in this sense, cannot be simply deleted or censored. Such diversity, because of what it is, needs some form of containment. If fan fiction were to give up the dialectic of containment and its own canonical frames, it would end up as a single, vast, nonnegotiable aleatory world and as socially meaningless free associations (note 17).
[6.8] Let me recapitulate. We should understand canon not as any particular fact, story line, or set of characters nor as an object, but, more flexibly, as the text's (sometimes self-chosen) containment practice that is invoked by the perception of superabundant potential, even as such potential can present in different ways and though different media in different historical contexts.
[6.9] Next I offer some brief examples of how this revised perspective on canon works in practice in the context of early orality. This has some useful implications for the idea of canon in fan fiction.
7. Canon and surplus in early Greece
[7.1] It is easy to observe orality from inside the narrow perspective of canonical early Greek epic, from inside what is now the written text of Homer's poetry. This, of course, is not what we are looking for. I will therefore say nothing of such canonical work except to note that it relies on the systematic repetition of hundreds, perhaps thousands, of ready-made traditional set phrases, known as formulae. These are well preserved in the written text of Homer, but their provenance is unmistakably oral. A large body of scholarship has persuasively argued that such repetition is inherently associated with the oral-traditional origins of Greek epic poetry: it makes it possible for a singer to compose long, well-formed poems in the process of performance without the use of writing (Russo 1997). This idea underpins my analysis.
[7.2] Where, then, would we find the evidence of oral noncanonical surplus? To have survived, it must obviously be written, yet I suggest that it exists in writing whose ontology is of a special sort: the writing of the everyday. Although relatively little of such writing from early Greece actually survives, it is the very everydayness of the objects in question that affirms their abundance beyond the absence of record.
[7.3] Our first piece of evidence is a short performative inscription—most certainly not the kind of writing associated with long, canonical epic poems. It appears on one of the best-known material everyday objects from archaic Greece and is one of the earliest examples of writing using the Greek alphabet in the classical world. The object was produced at a time when levels of literacy were low. Its text, although written, can thus be safely assumed to relate to oral usage. The nature of this object and the content of the words, which directly refers to the object, further assure us of the performative character of the text.
[7.4] The object in question is a simple fired clay kotylê, or drinking cup, dated to the second half of the 8th century (750–700 BCE), recovered in 1954 from a burial site in the ancient Greek colony of Pithekoussai on the island of Ischia, just off the Bay of Naples on the west coast of Italy. This vessel is commonly known as Nestor's Cup (Bartoněk and Büchner 1995) (figure 1). The cup is inscribed with the text illustrated in figure 2.
Figure 1. Nestor's Cup, Ischia (late 8th century BCE). [View larger image.]
Figure 2. Nestor's Cup inscription (CEG 454), reading, "Nestor's cup <I am>, good for drinking. / Whoever drinks from this cup, him, instantly / the desire of beautifully crowned Aphrodite will seize" (translation by Faraone 1996, slightly altered). [View larger image.]
[7.5] Upon close scrutiny, the inscription reveals an extraordinary dialectic of precisely canonical and noncanonical discourse and what might be described as its acts of appropriation—its use, in fannish terms, of original characters and original plot, and its acts of poaching, which are of necessity also acts of acknowledgment.
[7.6] On the one hand, the cup is clearly not a canonical object. It is not a grand, public artifact. Its material and form suggest everyday use. The cup seems to have been placed in the grave as a funerary offering in a particular location. However, precisely for this reason, we can safely assume that it was a strictly local object, not meant in any way for general circulation or for public veneration (note 18). The cup's inscription is likewise not grand, canonical poetry. It comprises three lines, of which the first seems to describe the object itself ("Nestor's cup <I am>"). The last two lines comprise a conditional magical formula of a type well attested in other ancient epigraphic texts and that is always associated with everyday, noncanonical discourse, often directly erotic, apotropaic, or vindictive. Such magical discourse is not attested in the canonical poetry of Homer, for example, or, by and large, in other archaic poems. Nestor's Cup, in other words, is an object very close to the life and death of ordinary 8th century BCE local users in Pithekoussai.
[7.7] On the other hand, the cup's inscription declares this to be Nestor's Cup. Nestor is a great Greek hero in Homer's epic. Furthermore, the two last verses of the cup's inscription, though not epic in content, each comprise, with some lesser variation (and partly dependent on supplement) a line of hexameter—an epos, as this meter was technically known in ancient Greek, the metrical form of epic, of Homer's poetry, and the single most authoritative, canonical generic form in antiquity. Neither the person who inscribed these words nor his readers or anyone listening to the words spoken are likely to have been completely oblivious to the words' canonical epic context. As one commentator on epigraphic verse puts it: "It seems quite possible that the composers of epigrams were in principle trying to compose in the local dialect, but that they attempted to give their poetry a specifically elegant touch by means of borrowings from the epic. After all, they used the epic hexameter" (Trümpy 2010, 176).
[7.8] By being marked as Nestor's Cup, this clay vessel is thus almost certain to have invoked its canonical epic namesake, known from its description in book 11 of Homer's Iliad:
[7.9] a beautifully wrought cup which the old hero [Nestor] brought with him from home.
It was set with golden nails, the eared handles upon it
were four, and on either side there were fashioned two doves
of gold, feeding, and there were double bases beneath it.
Another man with great effort could lift it full from the table,
but Nestor, aged as he was, lifted it without strain. (11.633–37)
[7.10] The monumental golden object of this great hero is clearly not the little clay cup of an unknown man buried at Pithekoussai. Yet as Christopher Faraone, one of the cup's important modern interpreters, notes,
[7.11] In recent years there has been a growing consensus that this inscription alludes to the epic tradition as part of a sophisticated joke that either plays on a humorous comparison between Nestor's enormous drinking vessel in the Iliad (11.632–37) and the humble clay cup which bears the inscription, or toys in rather subtle ways with the reader's generic expectations about proprietary inscriptions or conditional curses. (1996, 78)
[7.12] The Pithekoussai cup, though bearing a written inscription, is, as we have already suggested, early enough to exist in a largely oral environment. The canonicity of Homer's monumental epic has not silenced it. Indeed, the function of the cup depends on such canon and on some version of it. The canonical text may have been available in a more fixed, written form or, more likely, in more fluid, sung performances. The man who will have raised this cup to his lips is likely to have known the epic tradition, and if he did not, he is unlikely to have asked, "Nestor who?" and admitted his ignorance. The cup, both object and text, thus attest to an essential dialectic between canon and explicitly noncanonical local meanings and functions. The text poaches the Homeric source and character of Nestor, but it appropriates his role and transposes it onto a new context and the activities of an original character (in the fan fiction sense): the unnamed and far more humble real-life 8th-century symposiast. The cup reuses the source material for a purpose clearly neither intended nor envisioned in the canon, in "the story as told by the original author" (note 19). As de Certeau famously says, "Everyday life invents itself by poaching in countless ways on the property of others" (1984, xi).
[7.13] My second ancient example is again an unassuming early Greek everyday object, though it is somewhat less well known (except to fans of ancient epigraphy). This object too is far removed from monumental canonical texts, yet clearly it is has not been obliterated by the canon, and it is clearly dependent on essential textual poaching and a dialectical relation to canon. And again, although not many objects of its type are preserved from so early in Greek history, its everyday nature is proof that it represents a superabundance of similar, common objects.
[7.14] The object in question is a small (approximately 42 centimeters in height and width), plain funerary monument from the ancient Greek city of Kamyros on the island of Rhodes, dated to about 600–575 BCE (figure 3) (note 20).
Figure 3. Kamyros monument (IG XII 1.737). [View larger image.]
[7.15] The monument bears a short text inscribed in boustrophedon (bidirectional writing) on both sides of the stone. The words produce, in somewhat free metrical terms (as often elsewhere in epigraphic verse), an elegiac couplet, a common form made up of one hexameter line and a second verse comprising two half-lines of hexameter: "This tomb I, Idameneus, made, in order that I might have fame. / May Zeus make whoever harms <it> utterly accursed" (translation following Faraone 1996, 81, with minor alterations).
[7.16] Again we find a complex dialectic of allusion and transvaluation that at this period of Greek history is still on the oral cusp of literacy. This text, like our earlier example, comprises what we might describe as a fannish act of textual poaching that is directly comparable to contemporary fan responses, even as its historical, medial, linguistic, and material attributes are quite different.
[7.17] We have no idea who the Rhodian Idameneus was. His monument was found by the roadside with no indication of context. His name is poorly attested elsewhere in the epigraphic record (although, curiously, there is another early Rhodian cup inscribed with this name!). The monument is a modest, undecorated object—clearly not a canonical memorial. As in the first example, the magical curse in the second line of the text attests to common, everyday practice. Yet the metrics of the verse and especially the first line, a hexameter, again strongly mark its generic alignment with canonical heroic poetry. This is most clearly attested in the words "in order that I might have fame," which conclude the first verse (note 21).
[7.18] Fame (kleos) is a rather important Greek word—crucially so in canonical archaic heroic tradition. Fame is the very raison d'être for the canonical hero. In Homer, the hero's response to the transience of life and to ineluctable mortality depends on the defiant performance of great martial deeds, often, paradoxically, with lethal consequence. The hero's excellence is rewarded with prizes (tripods, captive women, and so on), honor, and fame. This fame is then augmented, disseminated, preserved, and made imperishable (aphthiton) in song, which allows the hero to get as close to the gods as a mortal man can (Schein 1984; Nagy 1999). In Homer, a prominent example of this idea can be found, for example, in Achilles' words as he lays out his choice of fate between the prospect of a safe return from Troy and the glory of heroic exploits (Iliad 9.413). If he chooses the latter, says Achilles, "My return home is gone, but my fame shall be imperishable." In another instance, in book 10 of the Iliad, as the Greek warriors are suffering, Nestor (the same man whose great cup we have just discussed) contemplates the possibility of a great deed of reconnaissance and the godly fame it may bring: "Could a man learn this, and then return again to us / unhurt, why huge and heaven-high his fame would be" (12.211–12).
[7.19] In archaic epic poetry, thought is never separate from material form and metrical structure. The most important ideas are always embedded and repeated in traditional elements of diction. These are the elements described above as formulae. They are the most distinct mark of the orality of early epic traditions. The point—and here I must apologize for the technical nature of my comment—is that in Homer the idea of fame is most prominently articulated in distinct formulae that (in the original Greek) comprise a collocation of the noun fame and the verb to be anchored at the end of the metrical line. Such metrical-grammatical patterns function as the formal material marker of the idea, in this case of heroic fame, in the same way that in a Wagnerian opera a leitmotif marks a particular character or theme. These Homeric metrical-grammatical markers of the epic idea of fame are also exactly, formally articulated in the words and usage of the Kamyros inscription: "in order that I might have fame" (cf. variants: Iliad 7.451, 458; 10.212; 22.514; Odyssey 1.298; 4.584; 9.264; 18.255; 19.128). Our Rhodian Idameneus may be a private ancient epigraphic blogger but, by the very structure of his words, he invokes canonical heroic fame, which he applies to his own purpose.
[7.20] Idameneus' monument invokes not merely the abstract idea of canonical epic fame but also his specific canonical mythological namesake, one of the epic tradition's well-known characters, the great hero Idomeneus. The name of this famous warrior prominently appears in a long list, known as the Catalogue of Ships, of the members of the Greek armies who fought at Troy (Iliad 2.645–47): "Idomeneus the spear-famed was leader of the Cretans, / those who held Knosos and Gortyna of the great walls, / Lyctos and Miletos and silver-shining Lykastos."
[7.21] Although not a Rhodian, Homer's Idomeneus is geographically and poetically a south Aegean Doric neighbor from the island of Crete. Oral abundance often leads to multiple versions of facts. It is telling that in the Catalogue of Ships, immediately after listing Idomeneus' Cretans, Homer does indeed describe the Rhodian contingent and names their most important city, Kamyros, in highly similar formal terms (2.653–56) (note 22).
[7.22] We have here another example of a local anonymous figure, an original character (in the fannish sense), Idameneus, who has poached and appropriated the discourse and values of canonical epic poetry as well as the name of Homer's Idomeneus, making use of these to suit his own ends and agendas. Without Homer and canonical epic hexameter, it is almost impossible to explain the inscription's diction or indeed its force. Idameneus' inscription is written, but it is deeply embedded in traditions of oral discourse; at this early period in Greek history, we are still dealing with predominantly oral societies. Whether there will have been written texts of Homer or not, canonical heroic poetry will have been disseminated largely by oral performance. It is unlikely that there will have been many written texts, let alone full texts of the canonical Iliad, in circulation (Janko, forthcoming). What is important for our purposes here is not the literal medium, be it writing or voice, but the wider historical contexts of production and reception. We are dealing with poetry that on the one hand embodies the canon but that on the other hand attests to prolific, subversive local practice. Anyone who read or heard Idameneus' short funerary epigram will have also heard the resonance of its canonical traditional epic echoes or else will probably have been too ashamed to admit that he had not.
[7.23] Greek hexameter verse inscriptions from the 6th century BCE and earlier are not numerous, but the few that have survived can, I suggest, be analyzed in the same way as my two examples (note 23). Furthermore, it is easy to demonstrate that most, if not all, are themselves largely local and everyday, and in this sense noncanonical, while they simultaneously resonate with the canonical language and thought of monumental heroic epic. As everyday objects, they attest to a much more prolific discourse. Indeed, what accounts for their scarcity is not the transience of orality or an annihilating effect of writing. Rather, it is the fragility of the material substance of their writing—clay and stone, which can be broken or taken away.
8. Early Greek canon and fan fiction
[8.1] I have not discussed, nor, in what little remains of this essay, do I propose to discuss, examples of the canonical dialectics of fan fiction. Many other scholars have done so recently, in detail and to good effect (Aardse 2014; Hellekson and Busse 2014; Siuda 2014; Hellekson 2015; Lindgren 2015). What I do hope to have shown by my comments on the canonicity of early Greek epic and by brief accompanying reference to key aspects of fan fiction is how similarly these two distant ancient and modern domains operate in relation to canon. I also hope to have hinted at how similar the determinant contexts of canonicity and general conditions of cultural reception and production in both these cultures are, despite vast formal differences. I would not be so foolhardy as to try to argue any of this in full in this short essay. However, it seems to me safe to suggest that paradoxically, despite obvious material, medial, economic, political, technological, and other differences, in archaic Greece we can find a directly comparable kind of small-community localism and fluidity of association that is characteristic of contemporary digital cultures and of fan fiction. It likewise seems to me safe to suggest that such fluid localism is significantly different from the highly literate, structured, and centralized social formations of, say, Virgil's Augustan Rome or the highly literate, bureaucratized national communities of emergent European nation-states, capitalism, and modernity in the 18th century, or the high modernist literary cultures of Paris and London in the early part of the 20th century. If such broad observations have any truth to them, and if there is any substance to the pointed arguments I have made here, then perhaps we will indeed have walked, if only for a short distance, the middle road I hoped for, learning something about both the historical uniqueness of canonical dialectics and about their pervasive universal necessity.