This book situates witchcraft drama within its cultural and intellectual context, highlighting the centrality of scepticism and belief in witchcraft to the genre. It is argued that these categories are most fruitfully understood not as static and mutually exclusive positions within the debate around witchcraft, but as rhetorical tools used within it. In drama, too, scepticism and belief are vital issues. The psychology of the witch character is characterised by a combination of impious scepticism towards God and credulous belief in the tricks of the witch’s master, the devil. Plays which present plausible depictions of witches typically use scepticism as a support: the witch’s power is subject to important limitations which make it easier to believe. Plays that take witchcraft less seriously present witches with unrestrained power, an excess of belief which ultimately induces scepticism. But scepticism towards witchcraft can become a veneer of rationality concealing other beliefs that pass without sceptical examination. The theatrical representation of witchcraft powerfully demonstrates its uncertain status as a historical and intellectual phenomenon; belief and scepticism in witchcraft drama are always found together, in creative tension with one another.
Whatever its flaws, Old English literature continues to rebuke the humanist narcissism that denies non-human animals possession of symbolic language. Like us, the early English knew that in singing, birds speak. As Susan Crane notes, at least one strain in medieval Western European thought held that birds composed ‘a society with a metaphoric relation to human society, in which birdsong fills the function of human language’, 2 and recent critics have heard welcome eruptions of interspecies intercourse in the avian voices of Old English literature. 3 In this vein of ecocritical optimism, I too read in the bird language of Beowulf a profound moment of interspecies connection. But I argue that within Beowulf the human is excluded from and indeed denigrated by the intimacy of wild creatures; when birds gossip about human corpses, this intimacy thematises the breakdown of socially embedded human knowledge.
Beowulf is an ideal site for those creatures and readers drawn to dire human straits, since far from a celebration of heroic achievement, the poem is a relentless chronicle of human failures. The most abject and notorious of these include queens’ failures to weave peace; fathers’ failures to protect and avenge their sons; Danes’ failures to defend themselves; pagans’ failures to communicate with the divine and achieve salvation; human failure to control objects and make things work, as in the cases of both swords and gold; Beowulf's failure to ensure peace and security for his people; and the ultimate failure of the heroic ideal of lordship, which impossibly requires both wise statecraft and martial recklessness. Beowulf's poetics of human disappointment and disaster deserves endless study. Mary Kate Hurley suggests that the avian encounter near the poem's end might provide a form of recuperation to Beowulf's ‘endlessly failing human communities’. 4 This chapter, while inspired by Hurley's provocation, strikes a more pessimistic note. It focuses on one crucial form of human failure in Beowulf – the failure of human knowledge enclosed in the homosocial bond of intimacy through which communication passes – and reveals that this bond endures in the realm of animal intimacy, where raven, wolf, and eagle triumph over the human.
This chapter begins with a moment of human loss. The murder of Æschere, the Danish king Hrothgar's runwita (knower of secrets, elsewhere interpreted as confidant, soul mate, and reader of runes, a point to which we shall return), severs a crucial past relation within which human meaning is made. As I shall argue, and as Thomas Meyer's experimental translation makes particularly explicit, when the Avenger 5 mutilates and kills Æschere, the total subordination of beasts is part of what is lost. Hrothgar's move to recuperate this loss through the conventional idiom of mourning – vengeance – involves, as an addendum to the eulogy, an anthropocentric remaking of the landscape as oral map. In the second movement of this chapter, I discuss a future intimacy, that of raven and eagle, in which beastly knowledge triumphs over human capacities for translation, for meaning making, for vengeance, and even for bodily integrity. In keeping with its focus on translation, the chapter reads translations of Beowulf as interpretations and as literature in their own right.
How animals convey meaning
Critical animal studies, the field that deconstructs human exceptionalism, has proven extremely congenial to Old English studies, prompting questions about non-human and post-human subjectivity, the maintenance of the human–animal boundary, and the costly constitution of the human in Anglo-Saxon culture. These inquiries track a larger ‘animal turn’ within medieval studies 6 and an ecocritical strain within early medieval studies. 7 Birds have drawn particular attention within medieval animal studies: not only the avian figures of the Exeter Book, but also the Bayeux Tapestry birds, the arguing pair of The Owl and the Nightingale, Chaucer's debating birds, and other medieval English textual fowls are provocative figures of animal language and knowledge transmission. 8 Peggy McCracken notes that in Yonec, Marie de France positions the speaking hawk as a figure who can communicate with humans, and yet who retains a store of untranslated (untranslatable?) knowledge. 9 Bird language often retains this sort of remainder, as we shall see.
In medieval literature and visual culture, non-human animals convey meaning through diverse methods; for instance, as visual images alone they ornament (e.g. the ‘wyrmfah’ [serpent-patterned] sword [1698a]), 10 provide visual allegory in religious images, take on narrative function in visual storytelling (e.g. the beasts of the Bayeux Tapestry, or the birds of the Illustrated Old English Heptateuch), act as emblems for human groups or gods, and behave as graphs attached to phonemes (e.g., the u-rune, ur [aurochs or wild ox]). Animal bodies are, as Nicole Shukin has perceptively noted, the material substrate of human communication: 11 they provided bones and skin on which to write, feathers that carried ink, and at times ink as well. They were also messengers and informers – both fictive and real. And they provided the symbolic language, allegorical symbols, and characters for human-authored literature, including the word-stock for place and human names (e.g. ‘Dæghrefn’ [Day-raven] [2501b], whom Beowulf kills in battle, and ‘Earna Næs’ [Eagles’ Nest] [3031b], where Beowulf's failed war-band finds his and the dragon's dead bodies) and for kennings (‘ganotes bæð’ [gannet's bath, i.e. sea] [1861b]). Birds bear a disproportionate burden in this cluster of signifying roles, since they have a particular relationship to language: they are the messengers between humans, between non-human species, and between the mundane world and the divine. And they can have a particularly strong metonymic relation to violent death, especially in Beowulf, where the most tragic form of human death is described as ‘hrefne to hroðre’ (pleasure for the raven) (2448a).
Within this critical environment, it is worth paying a bit more attention to an avian speech-act at the end of Beowulf. Wiglaf's messenger to Beowulf's people, the Geats, announces Beowulf's death and predicts a brutal invasion once the Swedes hear of his passing. He concludes a harrowing series of predictions with avian speech: ‘se wonna hrefn / … / earne secgan, hu him æt æte speow, / þenden he wið wulf wæl reafode’ (the dark raven … will tell the eagle how he surpassed him in eating, when he with the wolf laid waste to the slain) (3024a–7). I argue that this moment of inter-avian intimacy denigrates not only the slain human body, the fighting and agential body reduced here to morsels of warm meat for the pleasure of birds, but also the intimate human relations that, in the oral culture of the poem, provide the conduit for human knowledge production.
Homosociality and knowledge
In elite homosocial cultures, including both the warrior culture of Beowulf and the monasticism that probably produced the text, love and knowledge are transmitted through the same intimate bonds. 12 David M. Clark demonstrates the importance of the homosocial bond in Beowulf and, in Old English literature generally, the ‘utter wretchedness of being alone … without a brother warrior’. 13 A friend or lord is necessary for material and martial survival (as many of the exiles in the elegies point out) and for the sharing of joy and sorrow; Beowulf seems to mean all of this when he tells his lord, uncle, and foster brother Hygelac, ‘Gen is eall æt ðe / lissa gelong’ (All delight still depends on you) (2149b–50a). 14 A companion is necessary, too, for that currency that relates to all other aspects of emotional and physical survival in an oral culture: knowledge. As Benjamin A. Saltzman observes in his contribution to this volume, evoking the multiple senses of intimare, intimacy is the medium for the culture's stories and values.
That knowledge occurs only in relation, 15 and that the most intimate form of human relation in this culture is (in theory) the elite homosocial pair, provide the conditions for the tragedy at the chronological heart of the poem: the death of King Hrothgar's ‘hæleþa leofost’ (most loved warrior) (1296b), ‘aldorþegn’ (senior thane) (1308a), and ‘deorestan’ (dearest) (1309a) companion, Æschere, at the Avenger's hands. We meet Æschere, killed in Hrothgar's hall after Beowulf's triumph over Grendel, only posthumously:
Ne frin þu æfter sælum! Sorh is geniwod
Denigea leodum: dead is Æschere,
Yrmenlafes yldra broþor,
min runwita ond min rædbora,
eaxlgestealla ðonne we on orlege
hafelan weredon, þonne hniton feþan,
eoferas cnysedan. Swylc scolde eorl wesan,
æþeling ærgod, swylc Æschere wæs. (1322–9)
(Don't you ask about happiness. For the Danish people, sorrow is made new. Æschere is dead – Yrmenlaf's older brother, my knower of secrets and my counsellor, comrade in arms, when we guarded our heads in battle, when armies crashed and struck the boars. As a man ought to be, a fine prince, so Æschere was.)
This royal grief, framed first as the entire people's, is also deeply personal to the king. Æschere's role in Hrothgar's life, as presented in these few lines, is capacious: he is the companion of his youth, dear friend, essential advisor. Hrothgar seems to describe him, indeed, as an extension of the lordly body, grieving, ‘nu seo hand ligeð, / se þe eow welhwylcra wilna dohte’ (now the hand that gave you every good lies at rest) (1343b–4). This is the ‘hand’ of a ring-giver, liberally distributing social goods, and thus behaves functionally as Hrothgar's own hand; in Hrothgar's grief, it is also the source of ‘welhwylcra wilna’ (every good). I have translated runwita more or less literally as ‘knower of secrets’, but that is not to say that I know what it means. The word appears only twice in Old English, here and in Guthlac, where it refers to the dying saint: ‘Ða se wuldormaga worda gestilde, / rof runwita’ (then this steward of heaven's glory, steadfast wise man, rested from speech) (1094–5). 16 It may mean that Æschere was a confidant, the keeper of Hrothgar's secrets – this role belongs to the intimacy of homosocial friendship. It may mean, additionally or instead, that he imparted secrets to Hrothgar – that their relationship was a vehicle for kingly wisdom. James Paz, exploiting the connections of both run- and ræd- to writing and knowledge, makes the intriguing argument that within this oral culture, Æschere was the man who read, guarded, and interpreted secret knowledge for Hrothgar and his people. 17
Hrothgar's outpouring of grief represents an intimate as well as a public loss. The moment of death is, as David Halperin notes, the moment of ultimate intimacy for a heroic couple (a hero and his companion), equivalent to consummation for a heterosexual couple. 18 There is no stigma to a grief so intense that it seems to erase all joy – Beowulf's location of his entire happiness in Hygelac recalls such a statement. Yet this is meant to be a death in battle, with one companion ideally dying in the other's arms. This pair has grown too old to make such a death likely, yet it is not impossible. Although the young Beowulf in fact survives his uncle in battle, the elderly Beowulf will die in heroic manner in the company of Wiglaf, who expresses an unfulfilled wish to die alongside him (2650–2). Grendel's mother, the Avenger, smashes the fantasy of a battlefield death and all of its comforting androcentric implications: homosociality, patriarchal kinship, control over animals as symbols, the legibility of blood, the meaningfulness of human suffering. Meyer's translation makes clear what is lost:
What we shared:
secrets stomping feet
battles arms swinging
slashed boar emblems blood
heads split wide open noise
A bitch's bare hands crushed
that model man. Somewhere now
a beast's lips suck the bloody
stump of Yrmenlaf's brother 19
The initial couplets, undisrupted by verticality or capitalization, enclose what is lost: the lack of differentiation, of what never had to be differentiated or separated previously. Here are the unnamed (and perhaps even unremembered) secrets shared by these two (as well as the capacity to keep and unlock secrets), the youthful limbs moving joyfully and in sync, the violence against the enemy, framed first as the destruction of his animal symbolism (the ‘slashed boar emblems’ on warriors’ helmets) and then as the graphic invasion of his body. Perhaps this final recollection tips the scale, the memory of broken heads too much in this moment of violence, for here the topography shifts. As the vertical interrupts, so too do the graphemes and poetics of hierarchy, the complete sentences, the need for plans, the rage. For here, in the reality of violence against the one Hrothgar loves, is also the reality of the monster – not a flat ‘emblem’ but an agent with a body and motives that Hrothgar tries to atomize, sexualize, chop away: ‘[a] bitch's bare hands’, ‘a beast's lips’.
When translating the poem's omniscient descriptions of the Avenger, Meyer renders her appropriately formidable and sovereign; for example, ‘se ðe floda begong / heorogifre beheold hund missera’ (the one who, predatory, had held the water's course for a hundred half-years) (1497b–98) is rendered as ‘that terrible mother of floods, / those deep regions’ guardian for a hundred seasons’. 20 In Hrothgar's angry, grieving voice, he translates her in terms of misogyny and dehumanization. This should not be seen as a literal translation, but as a succinct amplification of Hrothgar's claims that the Grendelkin are monstrous, bestial, wicked, and have no right to hold territory or avenge death. 21 Thus the vulgarity ‘a bitch's bare hands’ draws not only on the Old English lines this most directly translates (‘handbanan / wælgæst wæfre’ (a barehanded killer, a nimble, murdering visitor) (1330–1) but also on the Avenger's ambiguous relation to the human: apparently a woman, she lives among wolves and mothers a monster; she is, in some way, vulpine, a ‘brimwylf’ (sea-wolf) (1506a); her powerful body can slaughter without weapons, but she uses a knife. The capacious semantic range of ‘bitch’, with the particular capacity of that word to flicker across species boundaries and moral categories, encompasses the poem's association of Grendel's mother with maternity outside the patriarchial human order and with feminine perversity within it. And it makes explicit the alignment of Grendel's mother with the bestial world once so easily sliced through.
Grendel's mother gets away from Hrothgar, conceptually and physically; it is his friend, still whole in his mind (‘that model man’, ‘Yrmenlaf's brother’) who has in fact been reduced to parts (‘bloody / stump’). And it is necessary, too, to blame Beowulf, which is really to blame Grendel:
in revenge for that hard grip
of yours, the life torn from
that mother's son, that monster
who raided this hall & claimed
the lives of some of my best
warriors. I'm told two things
can be seen to prowl the nearby
borderlands, a male & female 22
Here, though, Hrothgar gets into ontological difficulties, for this beast-woman, monster's mother, acts from motives he understands well: she subscribes wholeheartedly to the culture of kin vengeance that governs the poem's other actors. 23 And so his enraged vision of her is not coherent; he overdetermines her as dehumanized ‘beast’, humanized ‘mother’, ambiguous ‘bitch’; he tries again: ‘two things’. And, as if he has no other knowledge about the Grendelkin, ‘a male & female’. Yet he has already revealed that he knows much more: their relationship, her mourning. This is where he turns from a description of his enemy to a plan of attack. Perhaps it is safer territory for him, although his account is little more coherent.
Intertwined with Hrothgar's intense personal grief is nostalgia for the symbolic mastery over the animal world that he enjoyed with his runwita. While some critics have reduced nearly all animals in medieval literature to the symbolic and typological, the ‘boar emblems’ of Hrothgar's elegy are rare in presenting animal as pure symbol: the very narrowest type of those beasts that Gilles Deleuze and Félix Guattari call ‘State animals’, images and archetypes around which a political identity coalesces. 24 They hearken to the strange description of Grendel's mother's horrifying powers as inferior ‘be wæpnedmen / þonne … / sweord swate fah swin ofer helme / ecgum dyhttig andweard scireð’ (to men's, when … a bloodstained sword, strong in its edges, cuts through the swine on an opposite helmet) (1284b–7). 25 Graphic, man-made, representative not of the beast it depicts but of the men who travel under its sign, this boar is made (from a Danish perspective, at least) to be cracked along with the shields it embosses, to be slashed with swords, to bear the humiliation and loss of the ones who carry it and to reflect, in humiliation, the glory of the men who violate it. Such violence can be inscribed on real animals, too, but while the monsters (voiceless but for their cries) are slaughtered, the bragging birds use their voices and their motility to triumph discursively over the dead bodies of humanity.
For the human characters of Old English literature, intimacy is a key site of knowledge production. Exiles, disoriented from the sources of meaning that governed their lives, produce new understanding in solitary relation to God. The heathen Danes, in unwitting one-sided relation with a divine that does not exist, cannot find the source of their misery. Hrothgar, failing to strike a mutual chord in Beowulf, is rendered mute. In bragging, gossip, and threats, histories and futures emerge. That different histories work in different relationships is not inherently a problem in Beowulf – on the contrary, as Rosemary Huisman demonstrates of Beowulf's retellings of the battle with Grendel's mother, it is a design feature. 26 The more pertinent issue for these characters, demonstrated poignantly in the Lay of the Last Survivor (2247–65), which narrates the loss of an entire people, is that when there are no relationships, history is threadbare and cannot bring forth a future. As Nicholas Howe has observed, ‘the gift of history’ – knowledge about neighbouring lands, ledgers of battle debts incurred and paid, and examples of successful and failed rule – is ‘more precious than gold or horses’ in the world of Beowulf. 27
The poem's ideology operates in part, as Saltzman and Alexandra Bolintineanu note, through the open discussion of secrets. In particular, this is how it makes its monsters. As Bolintineanu observes, Beowulf redeploys the ‘men ne cunnon’ / ‘god ana wat’ (people do not know / only God knows) trope, which occurs throughout Old English literature to refer to eschatological matters and other divine mysteries, to talk about the details of the Grendelkin's lives, thus, as she says, ‘signal[ing] the utter alterity of the monsters’. 28 Saltzman notes that secrecy is in fact key to the poem's entire epistemology, but ‘human agency is often inadequate both to the act of concealment … and to the act of discovery (Grendel is concealed by night; his traces can be examined only in the morning)’. 29 This secrecy allocates power not only to those who understand the unknown, but also to those who, standing outside the inner gates of knowledge, understand what is unknown or secret.
The secrecy of the landscape is, for Hrothgar, a crucial aspect of his territorial claims and strategies. Hrothgar turns immediately from his grief to present to Beowulf an ‘oral map’, in Daniel C. Remein's terms, 30 of the mere and a description of its inhabitants in which contradictions and lacunae in human knowledge are features, not bugs:
Ic þæt londbuend, leode mine,
selerædende secgan hyrde,
þæt hie gesawon swylce twegen
micle mearcstapan moras healdan,
ellorgæstas. Ðæra oðer wæs,
þæs þe hie gewislicost gewitan meahton,
idese onlicnæs; oðer earmsceapen
on weres wæstmum wræclastas træd,
næfne he wæs mara þonne ænig man oðer;
þone on geardagum Grendel nemdon
foldbuende; no hie fæder cunnon,
hwæþer him ænig wæs ær acenned
dyrnra gasta. Hie dygel lond
warigeað wulfhleoþu, windige næssas,
frecne fengelad, ðær fyrgenstream
under næssa genipu niþer gewiteð,
flod under foldan. Nis þæt feor heonon
milgemearces þæt se mere standeð;
ofer þæm hongiað hrinde bearwas,
wudu wyrtum fæst wæter oferhelmað.
Þær mæg nihta gehwæm niðwundor seon,
fyr on flode. No þæs frod leofað
gumena bearna þæt þone grund wite. (1345–67)
(I have heard the land's inhabitants, my people, the hall counsellors, say that they saw two such large border-walkers guarding the moors, foreign spirits. One of these was, insofar as they could know for sure, in the form of a woman. The other, wretched, walked the paths of exile in the shape of a man, except that he was bigger than any other man; in the old days, the people living there called him Grendel. They do not know his father, or whether any were born before him among hidden spirits. They hold this secret land – wolf-hills, windy headlands, dangerous fen-path – where the mountain stream flows down into dark abysses, water under the earth. It is not far from here in miles that the mere stands; over it hangs a frosty grove, a hard-rooted forest hangs over the water. There at night anyone can see a sinister marvel: fire on the water. Among the children of men there lives none wise enough that he knows the [mere's] bottom.)
Hrothgar turns here to the medium with which he is most comfortable – the (somewhat synthetic, in this case) bond of masculine intimacy – to communicate a litany of secret knowledge. Hrothgar speaks of ‘dyrnra gasta’ (hidden spirits) and ‘dygel lond’ (secret land), and gestures ostentatiously, as Bolintineanu observes, to the unreliable links in the chain of his knowledge: he has heard people say that one of the monsters was ‘þæs þe hie gewislicost gewitan meahton, / idese onlicnæs’ (insofar as they could know for sure, in the form of a woman), embedding the apparent fact of this monster's womanhood in the unreliability of second-hand knowledge and witness doubt, with room to distinguish between appearance and ontology. He notes that his informants do not know Grendel's fatherhood and lineage, ‘no hie fæder cunnon, / hwæþer him ænig wæs ær acenned / dyrnra gasta’ (they do not know his father, or whether any were born before him among hidden spirits), nor do they, or anyone alive, have personal knowledge of the mere's bottom: ‘No þæs frod leofað / gumena bearna þæt þone grund wite’ (Among the children of men there lives none wise enough that he knows the [mere's] bottom). 31
Yet Hrothgar demonstrates no uncertainty about who rightfully possesses this wild territory or the right to speak about it: ‘Ic þæt londbuend, leode mine, / selerædende secgan hyrde’ (I have heard the land's inhabitants, my people, the hall counsellors, say). By defining the ‘londbuend’ as ‘leode mine’, he lays an ironclad transitive claim to the land that the mysterious creatures (thus) wrongfully guard; by making the inhabitants’ testimony intelligible in the hall and by speaking this from Heorot, he affirms Heorot as the proper clearing house for information about them. The claim that it is impossible to trace Grendel's lineage, or discover whether he is first-born, thus becomes intelligible not as lack but as positive assertion: a negation of the Grendelkin's land rights. The archive that has its raw materials in the wilderness (the sights of monsters, of frightened animals, of eerie wonders) acquires meaning within the processing logic of the hall, where narrative undergirds possession, just as in the world of Beowulf all stories about the world of creation, of men and women, of non-human animals and of good and evil, acquire intelligibility only by retelling in the hall.
It is standard to assume that the ‘mere’, because of its name and its function as the home of outlaws (among other reasons), is located in a fen of the sort known to Anglo-Saxons; fens were productive and provocative cultural locations precisely because they were inhospitable to agriculture and thus large-scale settlement. As Andy Orchard and others have noted and as Christopher Abram explores in this volume, however, the Grendelkin's territory is a highly idiosyncratic wetland, with a diverse range of topographical elements, flora, and fauna (including, for instance, wolves and sea-monsters) that ‘can scarcely be harmonised’. 32 As such, it is hardly representational, yet its very unlikeliness seems to amplify the features that attracted hermits such as Guthlac and monastic communities to the East Anglian wilderness with which this passage is often associated. As Sarah Harlan-Haughey observes, the fens offer ‘a retreat from normative topographies, from normative human lifestyles’. 33
This map is what the ecocritic Bruce Braun calls an ‘environmental imaginar[y]’, a vision of the land divided into socially intelligible zones, including different forms of so-called wilderness. 34 As Braun notes, all wilderness is co-produced by the natural and the social; there is no pristine, uninhabited land until someone needs terra nullius. So, too, with the wilderness outside Heorot's gates, which must be wild – that is, its known inhabitants must be portrayed as incapable of holding territory, as Fabienne Michelet argues, for the purposes of the Danish succession's land claims. 35 The oral map is gloriously incongruous in its details, combining for instance uplands and lowlands in a single monstrous landscape. Yet it is nonetheless naked in its intentions – to place the monsters firmly outside the bounds of human civilization and to entice Beowulf to murder Grendel's mother, who has killed Hrothgar's companion. Thus, when Hrothgar gestures repeatedly in this passage towards things that no one knows, he is not humbly indicating the limits of human knowledge, but rather constructing a landscape of secrets for the benefit of a man who likes to find what is untouched and unknown, and kill it. The intimacy between Hrothgar and Beowulf, a shadow of the lost intimacy between lord and runwita, takes studied ignorance as its currency.
What the raven said to the eagle: birds in translation
If, as Saltzman observes, ‘Beowulf posits a past against the limits of human knowledge and the limits of its own narrative’, its animal world makes the future equally epistemologically threatening. 36 Bestial intimacy, expressed in shared enjoyment of and secret knowledge about human bodies, emerges among the ominous group of predators whose appearance signals the end of Geatish happiness. Raven, eagle, and wolf are, in Francis P. Magoun's trope, the ‘beasts of battle’ who show up in Old English, Old Norse, and arguably Middle High German literature where violent death is expected. 37 Although most scholars, including critical animal studies scholars such as Donna Beth Ellard, see these beasts primarily as symbolic figures in a human drama, Joseph Harris notes ‘the attribution of voice and even of language to the beasts’ in many of their iterations. 38 Yet this boast, as R. D. Fulk, Robert E. Bjork, and John D. Niles observe, is the only Old English ‘conversation’ among the beasts of battle, 39 and only its outline reaches human ears, at the triple remove of space, time, and voice. That is, the speech appears before it is spoken, presumably at a physical distance from where it is spoken, and in the voice of a human being who does not attempt to approximate it closely. This third remove, that of voice, is itself a multiple remove – the poet translates into Old English the words of the Geatish messenger, who is translating bird language into his own dialect of Old Norse. Indeed, he indirectly reports the raven's words – not actually the words themselves, but only the topic of the bird's conversation. Implied here is a pan-avian language; if the raven and eagle speak their own dialects, these are mutually intelligible. There is no attempt, as so often when humans describe birdsong, to approximate any specific sounds. Moreover, this difference in voice emerges in part from temporal difference, since the foretold future of the Geats is the ever-receding distant past of the poet, scribes, and readers; and as Robert Stanton observes,
Imagining the sounds made by nonhuman animals involves grappling with extinction, species development, domestication, and breeding practices over the course of a thousand to fifteen hundred years; it is an open question whether a domesticated pig in the year 900 sounded more like a present-day pig than any English speech from 900 resembles any English spoken today. 40
The details of the message are, in fact, fairly ambiguous, and even the simple paraphrase gives rise to a number of possible interpretations. In Roy Liuzza's verse translation, for instance, ‘the dark raven, / greedy for carrion, shall speak a great deal, / ask the eagle how he fared at his feast / when he plundered corpses with the wolf’. 41 Here, the eagle is the agent, the raven only his interlocutor, and the questioning implies that there is plenty of carnage for seconds. The ambiguity of wið (a preposition that can mean, among other things, both ‘with’ and ‘against’) demonstrates what else the human reader has no access to: the level of intimacy, the kind of relations between birds and wolves. Can birds and wolves speak over, maybe about, our dead bodies? Do they have mutually intelligible dialects? What else might they talk about over a meal of our flesh? Despite its potential polysemy, wið in Beowulf usually indicates an adversarial relationship (see lines 113, 144, 152, inter alia). 42 Yet even in the context of warfare or strife it can still take a cooperative meaning; for example, ‘[Grendel] sibbe ne wolde / wið manna hwone mægenes Deniga’ ([Grendel] would never make peace with any of the Danes) (154b–5), where it aids a negative reference to peace that underlines the hostile mood. Liuzza interprets the word as indicating cooperation between the scavengers. Can we, too, imagine the wolf and the raven as intimate partners in anthropophagous passion, egging each other on, worrying at the same corpses with beaks and claws and teeth, and finally resting to tally the number of still-warm bodies they consumed; or can we imagine them only as rivals, the raven snatching pieces of flesh from the wolf's jaws, the wolf swiping at the raven with her already bloodied claws?
Translators and critics have struggled to interpret the nature of these intimacies: raven and eagle, raven and wolf. The answer depends, in turn, on interpreting the speech so narrowly withheld from human comprehension. As Mary Kate Hurley notes, the verb reafian (plunder, lay waste), which attaches elsewhere in Beowulf to the actions of warriors, ‘suggests that human plunder can be equated – at least lexically – with the plundering of carrion eaters’. 43 Thus, in Meyer's experimental translation, the raven is making a battle boast: ‘The greedy raven will have tales / to tell the eagle of feats shared / with the wolf on slaughter's field.’ 44 In this interpretation, bird and wolf are allies in what to them are honourable feats on the battlefield, kills to be ‘shared’ between them, tales to be shared abroad. 45 J. R. R. Tolkien sees the meaning of wið quite differently, translating this line as ‘the dusky raven gloating above the doomed shall speak many things, shall to the eagle tell how it sped him at the carrion-feast, when he vied with the wolf in picking bare the slain’. 46 These options are not mutually exclusive: the raven and wolf may have met as strangers and before long become friendly competitors, engaged in tugs of war over bodies snapping at the joints. In times of abundance, it is possible to be generous to one's erstwhile enemies. In both cases, the framing of this speech as battle-boast rewrites the field of battle as an arena for beastly victory, with human bodies the spoils of war. This is knowledge that counts in the culture of birds of prey, worth sharing with a companion in the same way that Beowulf shares with Hrothgar his battle victories and his knowledge of the Heathobardish loot.
Ultimately, this brief and partial moment of knowledge-in-relation intervenes to devastating effect in the anthropomorphic world of the poem. The raven who crows victoriously over the slain Geats has accrued associations of battle-joy (1801) and the despair of unavenged mourning (2448) before this final appearance in the poem. In Gillian Overing's well-known argument establishing metonymy as the dominant sign system of Beowulf, one potent example is the glorious necklace that ‘weighs like a millstone on the narrative’; arriving at a moment of triumph, it links that moment to the bitterness of past and future by participating, along with the poem's other significant rings and cups, in an endless dance of mutual reference. 47 So too does the figure of the raven bring the joy of earlier victory into tragically ironic inversion at Beowulf's death. Yet this joy is not entirely inverted, for the raven's own jubilation, and that of his comrades, continues unabated; nor is the helplessness of grief abated, since it will pass into the possession of the surviving Geats. The beasts of battle provide a counterpoint of vitality and abundance to Geatish death and poverty. While the Geats will go hungry, the beasts of battle can brag about their gorging. Where the Geats were subjects of battle-knowledge, now they are objects. And while all of this is too distant from the human observer to be independently verifiable, the selectively omniscient voice of the poem affirms its truth-value: ‘Swa se secg hwata secggende wæs, / laðra spella; he ne leag fela / wyrda ne worda’ (So the man was telling evil news; he did not mislead much in words or deeds) (3028a–30). In other words, he did not mislead them at all.
Bird language in translation
It is difficult for humans to parse and taxonomize avian language. Harris lightly distinguishes between ‘voice and even … language’ in the case of the birds of battle, but the problem with any distinction between voice and speech, or between what Harris calls ‘sublinguistic cries’ and exclamations that count as speech (e.g., in English, ‘Oh!’) is that it is almost always impossible for an outsider to accurately determine; hence the very common descriptions of foreign speech as meaningless babble, equivalent to infant babble or animal noise (which are not necessarily themselves sublingual categories). 48 Moreover, it is based on a logocentric and Eurocentric view of language that pre-emptively evacuates sound quality (tone, pitch, accent) of the capacity to convey symbolic meaning, despite the evidence that it does so in every spoken language and, indeed, the fact that more than a billion humans speak a tonal language as their first tongue. Recent work on avian and particularly corvid intelligence has demonstrated that anthropocentric definitions of intelligence and of communication can operate only via logocentric tautology; (one particular strand of) human self-consciousness defines as intelligent only entities that engage in the abstract, rational processes that produce these exclusions. 49 As Laurie Shannon notes, in the Cartesian logic that classified non-human animals as machines, ‘Speaking only counts if it is speaking to us and in our language.’ 50 In this Enlightenment frame, the only thought that exists is the one that can be communicated to and understood by rational (educated, adult, white, Western, male, modern) human interlocutors. Nonetheless, avian scientists have demonstrated that birds, which are not evolutionarily close to humans, are intelligent to a degree that even humans can begin to comprehend. 51
I follow Ellard in noticing the importance of Ælfric's Grammar to discussions of bird language; I am particularly interested in how he has altered his source. He begins with a discussion of vox (speech or voice), the auditory basis of spoken language:
Omnis vox aut articulata est aut confusa. articulata est, quae litteris conprehendi potest; confusa, quae scribi non potest … ælc stemn is oððe andgytfullic oððe gemenged. andgytfullic stemn is, þe mid andgyte bið geclypod, swaswa ys arma uirumque cano ic herige þa wæpnu and ðone wer. gemenged stemn is, þe bið butan andgyte, swylc swa is hryðera gehlow and horsa hnægung, hunda gebeorc, treowa brastlung et cetera. 52
(All voice is either meaningful or mixed. What has meaning is called meaningful voice, as in I sing of arms and the man, I praise arms and the man. Mixed voice is what is without sense, such as the lowing of oxen and neighing of horses, barking of dogs, rustling of trees, etc.)
As so frequently in Ælfric's work, small alterations to his sources prove philosophically significant. 53 Ælfric begins with a Latin excerpt that derives from Isidore, and explicates it in Old English, borrowing heavily from Priscian, in a way that entirely changes both Isidore's and Priscian's categories.
Isidore distinguishes, as Karl Steel notes, between human vox articulata (‘articulated’ voice) and non-human vox confusa (‘confused’ voice); the first ‘scribi potest’ (can be written), while the second cannot. 54 Ælfric's main source for the Grammar, the Excerptiones de Prisciano (which includes material from Priscian, Isidore, Donatus, and others) renders Isidore's thesis as four separate but presumably overlapping types of sound: ‘articulata, inarticulata, litterata, inlitterata’ (‘articulate and inarticulate, those expressible in writing and those inexpressible in writing’). The Excerptiones explains these as follows:
Articulate speech is that which is pronounced in connected fashion, that is, joined with some meaning in the mind of the one who speaks, such as ‘I sing arms and the man.’ Opposite to this is inarticulate speech, which is uttered without the influence of the mind, such as whistles or groans. These cannot be written even though they can be understood. Speech expressible in writing is that which can be written without conveying meaning, such as coax and cra. Speech inexpressible in writing is that which can be neither written nor understood, such as a noise or bellow or the like. 55
Articulate speech is the speech of men; not only does it require ‘mind’ and intention, but its exemplary subject is men and their weaponry. More than in Isidore's accounting, and far more than in Ælfric's, such speech relies on a binary relation, a disavowal, for its definition; the ‘mind[less]’ sounds of inarticulate voice can be interpreted by the skilled listener, but there is no intellect crafting their message – they are sounds of the body. The third set includes nonsense syllables – manipulations of known phonemes without the intention to produce a message – and (as evidenced in the happy accident of coax) words in barbarian languages, whose meaning it is not necessary to know. The final category is the province of noise-making things. It is not entirely clear where bird and other non-human animal cries fit into this schema – as the carnal sounds of inarticulate speech? the transcribable nonsense of speech expressible in writing? the final category of abject sounds that might as well be silence? – and perhaps it is precisely this problem, among others, that prompts Ælfric's reworking of these categories.
Ælfric is perfectly happy to translate the Latin he incorporates into his Old English writings, even if this intimate mingling (as in ‘arma uirumque cano ic herige þa wæpnu’) creates redundancy for Latinate readers, although he does not always do so. Instead of translating Isidore's division between vox articulata and vox confusa as a matter of what can be written and what cannot, or glossing it in a way that preserves or extends the definition, he recasts the distinction as one of sense. Yet as Ellard notes, the flagrantly improper binary he chooses as gloss – ‘gemenged stemn’ (mixed voice), which also describes Ælfric's own hybrid writing, versus ‘andgytfullic stemn’ (meaningful voice) – functions as acknowledgement of animal language. 56 Although Ælfric describes hybrid speech as ‘butan andgyte’ (without sense), the very appellation indicates that it is meaningful – just not to humans, who cannot separate the static from the noise and parse the signals into information. It is, Ellard observes, ‘a semiotics that cannot be translated’. 57 That Ælfric renders the difference of animal language into a problem of translation, rather than of writing, may also indicate a recognition of the fragility of the Isidorean binary, for humans can and do transcribe birdsong. 58 It also, certainly, reflects the abbot's habitual preoccupation with the question of translation. To my mind, it is more appropriate to Old English literary culture, where the questions of what is known and what can be said are inextricably bound up with questions of translation.
Translatability is not, however, the only criterion for meaning as represented in Old English. Although Shannon understates the anthropocentrism of medieval European Christian culture, 59 there is still a crucial difference between these early medieval understandings of bird language and later reductions of bird language to the mechanical. It is not the case that sense that humans cannot understand is no sense at all. As I have argued, the speech of raven to eagle derives its sinister power in part from the fact that humans cannot directly translate its details. Its ‘mixed’ character – the intimate mixing of what humans can know and what they can never know – is the ideal form for its terrifying rebuke of the possibility of human survival through relation.
Ellard compellingly depicts The Phoenix and the Exeter Book bird riddles, the core of Old English literature's avian imaginary, in terms of a ‘vernacular ecosystem’ that enfolds birdsong/speech into Anglo-Saxon translation. This produces a hopeful reading that illuminates the connections between ‘avian intelligence’ and (post)human spirituality. 60 Similarly, Jonathan Hsy finds in Anglo-Saxon taxonomies of birdsong an interspecies ‘bilingualism’: ‘an intimate partnership that bridges species boundaries and language difference’. 61 Beowulf takes quite a different approach, approximating meaning but not sounds, and allowing the birds an ambiguity that triumphs over human comprehension.
Conclusion: the return of loss
Although Beowulf is, ostensibly, the portrait of a heroic culture that values homosocial intimacy, it is also incapable of imagining this intimacy without the shadows of failure and grief. We learn of the king's companion only in elegy; we perceive him only as a bloodied, disembodied head and in the image of the king's grief. Yet I do not suggest that we pursue this homosocial loss through the lens of what is now called queer mourning, grief for what has been rendered conventionally ungrievable. While it is not sufficiently significant to some modern queer readings to merit much analysis – Æschere does not, for instance, appear at all in Clark's or in Allen J. Frantzen's speculations on same-sex love in Beowulf 62 – Hrothgar's love is culturally intelligible and therefore this loss can be avenged with all available resources. Vengeance is primary, as Beowulf reminds Hrothgar, among what Judith Butler calls the ‘cultural conventions’ of mourning through which grief may be acknowledged and expunged in this world. 63 Hence the frustrated griefs of this poem, when those who cannot avenge find shame or depression their constant companions. In the act of violence itself, and in the acts of planning, coaxing, and bragging that precede it and boasting that follow, human bodily integrity is affirmed, homosocial cohesion restored, and supremacy over the wide and terrible non-human world temporarily regained.
Avian confidences bring the cycle of loss to a new beginning. When the raven and the eagle share a boast over human corpses, they are not only recapitulating the poem's paradigmatic moment of human helplessness, when a man past the age of strength watches the raven slowly delight in the body of his dear one and realizes the world is too much for him. Nor are they, to speak in more detached terms, only liberating avian life and language from their accustomed yoke of symbolism. They are also forging, and recounting, interspecies connections to which human intelligence is not privy, speaking a language that humans cannot even hear, much less accurately translate. They are making human flesh the material substrate of this non-human culture – our ravaged bones and bodies, our severed bonds, the fuel for animal knowledge and thought.