In the opening lines of Hárbarðsljóð, Þórr calls across a fjord to a ferryman (Hárbarðr or “Grey Beard,” a disguised form of Óðinn), only to encounter his own words altered and returned: Þórr kallaði: “Hverr er sá sveinn sveina er stendr fyr sundit handan?”Hann svaraði: “Hverr er sá karl karla er kallar um váginn?”2 (sts. 1–2)(Þórr called: “Who is that boy of boys who stands on the other side of the strait?” He answered: “Who is that peasant of peasants who calls over the bay?”)Finnur Jónsson (1920, 1:151) describes Hárbarðr's reply here as a “gækkende ekko” (mocking echo), choosing a vocabulary of rebounding sound that draws attention to the spatial gulf separating the two interlocutors: the sund (strait) or vágr (bay) invoked in these lines.For all that this turn of phrase is apt given the poem's imagined geography, the concept of echo has relevance across all those eddic poems that are cast wholly or mainly in dialogue form, which number at least ten according to Terry Gunnell's classification (1994, 191–2). In these texts, words, phrases, and syntactic structures are often repeated, in part or wholesale, and reframed across different speakers’ utterances. From one perspective, this is not a phenomenon specific to the dialogue poems: across the eddic corpus as a whole, devices of repetition and parallelism are present in a “restrained” but recurrent way, manifesting in refrains as well as shorter anaphoric or epistrophic motifs (Meletinsky 1998, 17).3 Such repeated structures and motifs can nonetheless be seen to form a distinct phenomenon when employed within a context of two or more speakers; in such situations, the continuity of the reused language is interrupted by clear disjunctions of voice, and interpersonally significant applications for repetition are made available, including affirmatory, persuasive, competitive, or subversive quotation of others.Previously, scholars have tended to contextualize the echo-replies of Hárbarðsljóð with reference to the wider tradition of organized competitive speech known as flyting, defined by Antje G. Frotscher as “an antagonistic, person-oriented, in some way formalized verbal contest” (2003, 4). Hárbarðsljóð has been categorized as a mixture of the Old Norse-Icelandic flyting genres of senna, consisting of sequenced provocations and mannjafnaðr (man-equaling), sequenced boasts.4 In her account of the poem, Carol J. Clover (1979, 127, 136) quotes Finnur Jónsson on the opening “gækkende ekko” and goes on to reflect that flyting in general operates according to a metaphorical understanding of “language [as] equivalent to ammunition,” such that verbal repetition is comparable to “firing back the enemy's own spear or arrow.”5 Marcel Bax and Tineke Padmos similarly identify Hárbarðsljóð’s “strategies of mirroring and surpassing” as “normal procedures in flyting matches” (1983, 153). Indeed, Earl R. Anderson, with reference to The Battle of Maldon, observes that “ironic verbal echoes are a conventional feature of heroic flyting” (1970, 199) and builds on Edward B. Irving, Jr.’s declaration that, in such battles, words “are deftly caught, ironically accepted, and sent back in a notable display of heroic wit” (1961, 460).The repetitious nature of medieval flyting itself has occasionally been contextualized amid wider landscapes of language use. Parallels have sometimes usefully been drawn between the repetitive style of eddic flytings and eddic wisdom contests; in both, “speakers must not only answer to the substance” of the question or insult, but “also prove their rhetorical skill by mirroring and surpassing the language and form of the initial volley.”6 Reaching beyond medieval contexts altogether, the repetitious dialogue of medieval flyting has also been compared to what is sometimes seen as a modern reflex of the same practice of verbal dueling: “sounding” in inner-city African American communities in the 1960s and early 1970s, defined as “the game of exchanging ritualized insults” and influentially analyzed by William Labov (1972, 306).7 Comparisons across these contexts are often illuminating, including with regard to the shared importance of “speech-linking” in medieval flyting and in modern insult games (Parks 1990, 114–5). Labov discusses how an initial sound “opens a field” and subsequent replies “must build upon the specific model offered,” whether through adopting “the substance” (the attribute discussed) or “the same surface form”—optimally successful sounds will then bring about “striking semantic shifts with minimal changes of form” (1972, 344–7, 349). Such principles have much in common with the kind of “mirroring and surpassing” critics have observed in Hárbarðsljóð (Bax and Padmos 1983, 153; Schorn 2016b, 280), and modern traditions of stylized competitive speech thus offer a helpful context for understanding certain features of Hárbarðsljóð and other eddic flytings.At the same time, it has also been noted that if medieval flyting and modern sounding can be spoken of as part of the same history, “it is at best discontinuous and disjunctive”—Leslie K. Arnovick argues for an intimate connection between “orality” and “the agonistic,” to the extent that “both flyting and sounding prove independent reflexes of something that may well be a linguistic universal” (2000, 38). Arnovick mentions “verbal echoes” (e.g., “mirroring” and “surpassing”) as distinctive of flyting, but is more generally concerned with the history of the “English agonistic insult” (2000, 30, 38). Her concept of the “agonistic” is inherited from Walter Ong, who sees it as kind of “adversativeness” present across human societies in diverse forms, including as play or contest (1981).8No claim for universal agonistic tendencies is made by the present paper, but I do intend to widen the lens through which we view the “verbal echoes” of Hárbarðsljóð, as well as those of another eddic flyting, Lokasenna, in a manner that to some degree complements Arnovick's project. I aim to achieve this by drawing on a number of linguistic studies that have, in recent years, tackled the phenomenon of cross-speaker echo in a variety of speech contexts. Eddic poems are, of course, highly wrought literary creations, metrically and formally complex, and not spoken extemporaneously as conversational discourse, but, equally, literary texts can never be fully separated from more day-to-day forms of communication.9 The field of speech pragmatics has much to offer in its demonstration of the sheer versatility of cross-speaker echo as a feature of dialogue, and the wide range of effects it can have. Engaging with this body of scholarship allows us to appreciate how the potential implications of verbal echoes extend beyond antagonism, and beyond even agonism, into a complex range of adversative, collaborative, and cooperative stances.Verbal mirroring in flyting can, in particular, be understood as a conspicuously formalized version of a much more ubiquitous linguistic phenomenon, one which has been found to underlie practices of speech exchange far more widely, including in casual and conversational contexts. This is what John W. Du Bois (2014) calls “dialogic syntax,” and it is notably a collaborative enterprise between speakers. Dialogic syntax comprises “the linguistic, cognitive, and interactional processes involved when speakers selectively reproduce aspects of prior utterances, and when recipients recognize the resulting parallelisms and draw inferences from them” (Du Bois 2014, 366). “Diagraph” is his proposed term for counterpart structures that are shared across utterances, extending beyond the unit of the sentence. The following example of a diagraph, organized into columns to highlight parallelisms, is selected by Du Bois; this exchange follows a critical remark that the first speaker, Joanne, has just made about her mother: JOANNE;it’skind oflike^youKen .KEN;that’snot at ^alllikemeJoanne .10Rather than choosing from any one of a number of other options for formulating his reply (such as simply “‘No,” or “I disagree”), Ken produces a response that selectively reproduces features of Joanne's utterance. As identified by Du Bois, these include pronouns (“it”: “that”; “you”: “me”), proper names, adverbial modifiers (“kind of”: “not at all”), and morphologically identical units (“like”: “like”; “’s”: “’s”), as well as co-reference (“you” and “me” referring to Ken, “it” and “that” to the quality associated with Joanne's mother), and the copular predicative construction (X is Y) at the phrasal level. Despite all these points of consistency, Ken makes a diametrically opposed claim. Structural parallelism here enables subversion, and this is particularly clear in Joanne's and Ken's respective uses of the vocative: Joanne's use of Ken's name is necessary to specify her addressee (the conversation included a third participant), but Ken's use of Joanne's name serves no such clarifying function: “vocative Joanne takes cover as tit for tat, while actually dripping with irony” (Du Bois 2014, 363). In such ways, parallelism can shape utterances in which “the second speaker's meaning is parallel, opposed, or even orthogonal to that of the first” (2014, 360), explored at length by Du Bois in his work on “stance-taking” in conversation (2007).Paralleled structures for speech thus enable a spectrum of many possible implied stances. The theory of dialogic syntax offers a way to approach the patterned dialogues of eddic poetry in a manner that both encompasses and exceeds the kind of speech-linking previously understood to characterize flyting. As will be seen, competitive or confrontational frameworks for speech do not preclude the implication of other kinds of stances: parallel, askance, or “orthogonal.” Appeals to shared knowledge, the seeking of information, and attempts to persuade or coerce may all be implied through verbal echoes within the broadly antagonistic framework of eddic flyting.Other recent research into cross-speaker repetitions in conversational contexts has light to shine on the possible workings of echo in Hárbarðsljóð and other eddic dialogues. Some researchers have focused specifically on the effects of echo-answers and echo-questions. In a study spanning fourteen languages, Enfield et al. have recently observed that echoing (or “repetition-type”) answers, in response to polar (yes-no) questions, may “be better suited to contexts in which the answerer aims not only to confirm the proposition that has just been put on the table by the questioner, but to claim a degree of thematic agency, or independent interest over that proposition”; this could be because the domain of the proposition falls within the second speaker's “special realm of knowledge” (2019, 286, 292; after Heritage and Raymond 2005; 2012). Repeating parts of the question may thus enable Person B to “[push] back against the implicit claim by the questioner to primary interest in thematizing this proposition” (Enfield et al. 2019, 286).This complements previous research into “echo questions,” when a questioner repeats all or part of their interlocutor's previous utterance. Scholars have focused on the ambivalence of the device, and the potential challenges it poses: echo questions “are not used only for confirmation of the words or sounds of an utterance, but may also draw attention to the absurdity of an underlying proposition” (Channon, Foulkes, and Walker 2018, 158; after Blakemore 1994). This phenomenon has serious implications for contexts such as Language Analysis for the Determination of Origin (LADO) interviews with asylum seekers, as “echoes may . . . be treated as repair indicators indicating a problem with the content of prior talk” (Channon, Foulkes, and Walker 2018, 162). Highly subtle inferences can therefore be drawn from verbal echoes in conversational contexts. Although the implication of “a problem” and the need for “repair” can be understood as broadly antagonistic, reducing it to such would obscure a great deal of nuance. The same would be true of the suggestion of a “special realm” of knowledge—such a claim may signal antagonism and a kind of competitive impulse, but that is not all it does. As will be seen, expanding the vocabulary with which we describe the echoes of Hárbarðsljóð and Lokasenna has the potential to elevate scholarly discussion away from the “mirror and surpass” device understood as pure verbal competition and/or the assertion of a hostile stance, and toward appreciation of potentially more complex dynamics of speech exchange captured in these texts.Given that these poems may have been designed for embodied performance, the complex implicational opportunities presented by devices of echo could conceivably have been taken up by oral performers. Arguments for the dramatic nature of eddic poetry have most influentially been advanced by Bertha Phillpotts (1920) and Terry Gunnell (1994). Other scholars perceive the poems as “literary fictions of a human poetic voice,” such that “a human audience for these fictions completed the locutionary circle” (Clunies Ross 2016, 20).11 In either scenario, echoes between speakers create opportunities for a range of intonational possibilities, whether actualized or imagined.As understood by Du Bois, prosody plays a crucial role in the development of connected conversational discourse. Stress and intonation provide a rich source of parallels and contrasts between utterances and between the implied stances of speakers; in the Joanne/Ken example above, both utterances consist of a single intonation unit, and this basic similarity is accompanied by both other prosodic affinities (such as in the use of final intonation) and differences (such as the placement of the primary accent, marked with ^) (Du Bois 2014, 362).12 Du Bois's diagraphs are intended to capture “an index of the voices heard, in all their prosodic particularity” (2014, 396). When speakers replicate parts of each other's utterances, these repeated features create space for significant discrepancies not only of lexis and syntax, but of potential voice quality and tone.How precisely these opportunities for voicing might have unfolded in Old Norse-speaking contemporary contexts is up for debate, though largely irrecoverable. Introducing a survey of twenty languages, Daniel Hirst and Albert Di Cristo (1998, 6) identify the phenomenon of intonation as pluriparametric, crucially involving, on a physical level, “fundamental frequency, intensity, duration, and spectral characteristics,” corresponding loosely to the auditory experience of “pitch, loudness, length and timbre”; among these, fundamental frequency, or the perception of pitch, is generally acknowledged to be the primary parameter.13 Rhythm, or “aspects of temporal organisation,” can then be understood as reflected in the metrics of intensity, duration, and spectral characteristics (Hirst and Di Cristo 1998, 4; after Crystal 1969). Old Norse is sometimes posited to have been a pitch accent (or tonal accent) language, in the manner of modern Norwegian or Swedish; in such languages, the use of pitch as a cue for emphasis or expressivity on the level of the phrase or sentence can interact with word stress in complex ways.14 Previous questions regarding the role of pitch in early recitations of Old Norse poetry have been raised with reference to skaldic texts, especially as scholars have wondered how the poetry's characteristic parenthetic clauses might have been “set off from the rest of the stanza” to aid the comprehension of listeners (Gade 1995, 189). Hollander suggests that rather than through the use of a “different pitch,” the parenthetic constructions could have been distinguished by “a pause before and after” (1965, 636). No consensus exists in this area, and it seems likely that none will be reached on how precisely the dialogue of eddic poetry was voiced, and what kind of intonation might have helped to pick out key parallels and divergences. I wish only to suggest here that the paralleled dialogic constructions of the written texts may have evoked possible variances of intonation, which could have been actualized in performance (or in the literary fiction of an embodied voice) to signal different kinds of stance.Hárbarðsljóð will be considered here alongside Lokasenna: another mythological eddic dialogue strongly combative in tone, with which Hárbarðsljóð is often compared.15Lokasenna differs from Hárbarðsljóð in several key regards, most strikingly in its number of speakers. While Hárbarðsljóð constitutes a two-person flyting, Lokasenna consists of a series of flyting-like exchanges between Loki and sixteen other speakers. Nonetheless, both share formal arrangements typical more generally of eddic dialogue. Hárbarðsljóð is highly metrically various but joins Lokasenna in making use of ljóðaháttr (song meter), common in eddic poetry when non-narrative direct speech is involved (Phillpotts 1920, 26; Poole 2005, 268–9; Gunnell 2016, 97–8; Schorn 2016b, 279) and largely employed in “mythological and sententious” poems (Fulk 2016, 262). In the Codex Regius and AM 748 (in Hárbarðsljóð’s case), these two poems are also among those marked with marginal notation identifying different speakers (Gunnell 1994, 208–11), suggesting a possible connection with dramatic performance. As will be seen, while the conversations in Hárbarðsljóð and Lokasenna are largely antagonistic in nature, echoing responses are often used to articulate something more “orthogonal” than might be expected, with implications for both flyting and verbal battles more widely. In their use of such echoes, Hárbarðsljóð and Lokasenna may ultimately be understood as more continuous with other modes of speech exchange than is often supposed.16Additionally, each poem presents its abundance of interpersonal verbal echoes within a different set of “reported circumstances,” a dramatic quality noted by Ursula Dronke (1997, 395). Verbal exchanges in Hárbarðsljóð take place across a sund (sound, strait), and in Lokasenna during a sumbl (feast, gathering). The poets of these texts are notably interested in referencing these physical situations, including the way sound resonates within each; ultimately, they create a framework around their cross-speaker echo that draws attention to the movement of sound across space. Scholarly language used to describe competitive speech and dialogic syntax has already been found to invoke physical space in subtly metaphorical ways. Labov describes how an initial utterance “opens a field” in sounding, while Du Bois argues that “the diagraph frames syntactic affordances for the discovery of common ground between interlocutors” (Labov 1972, 344–71, 349; Du Bois 2014, 397). As will be seen, the poets of Hárbarðsljóð and Lokasenna are both alive to the ways in which attitudinal “stance” can feed into and draw upon physical “stance,” in the sense of a “standing-place, station, position.”17Highly metrically irregular, this text has long been something of a “puzzle” to critics (Arnold 2014; see also Swenson 1991, 25–6). Recent critical attention has focused almost exclusively on the dialogue's identity as a flyting, but the details of how exactly the text participates in this genre have been the cause of considerable disagreement. Clover (1979, 138) sees it as a comic genre parody, loaded with a “superabundance of flyting clichés” and committed to subverting generic conventions, particularly through the interventions of Hárbarðr as he boasts about unconventional topics and runs circles around Þórr's “intellectual inadequacy.”18 Bax and Padmos (1983, 157) have disagreed that the poem is parodic, seeing it instead as centring on “two competent language users” co-producing a relatively sophisticated piece of antagonistic dialogue. Arnold (2014, 7) argues that a “critical stalemate” has been produced by these two accounts, and proposes a new interpretation that incorporates much of Clover's sense of the poem's comic tone, supplements this with Bax and Padmos's sustained distinction between the use of senna and mannjafnaðr, and notes that ritual verbal battles may typically possess a kind of comedy. Arnold then reunites these more recent forays in Hárbarðsljóð scholarship with the previous center of scholarly discussion around the poem: the sociopolitically charged identities of Þórr and Óðinn. Dating the poem late, after the conversion period, and possibly from Norway, Arnold contends that it engages with the relative status of Þórr and Óðinn worship in the late Viking Age: the poem's references to social rank (especially in st. 24), and its “apparent preference for the intellectually sophisticated Óðinn,” could then be understood as reacting to the persistent popularity of Þórr worship amid certain groups in western Scandinavia in the late tenth and early eleventh centuries, problematic from the perspective of newly Christian aristocrats and monarchs, many of whom either previously worshipped Óðinn, or were descended from those who did (Arnold 2014, 18). Elsewhere in eddic mythology, Þórr and Óðinn appear as complementary figures, happy to collaborate as a part of wider strategies: Hárbarðsljóð can therefore be found to dramatize, at points, “the breakdown of their relationship” (Arnold 2014, 22).This theme of interpersonal and symbolic distance between Þórr and Óðinn can be newly appreciated as accentuated by inter-speaker repetition and parallelism. We have already seen an example of these devices serving such a function in the opening lines of the poem, as Hárbarðr replicates Þórr's syntax, but categorizes him as a “karl karla” [peasant of peasants]. Repeated syntax and lexis is used here to measure out social difference. Moments of echoed speech occur elsewhere in Hárbarðsljóð in conjunction with sensitive issues of identity. These sometimes involve issues of rank and sometimes make oblique reference to shared personal history and the (thwarted) possibility of collaboration. The latter tensions surface, for instance, in the central mannjafnaðr section of the poem, following a salacious description by Hárbarðr of his womanising exploits, typical of his own particular mode of self-aggrandizement:19Hárbarðr kvað: “Liðs þíns væra ek þá þurfi, Þórr, at ek helda þeiri inni línhvítu mey.”Þórr kvað: “Ek mynda þér þá þat veita, ef ek viðr of kœmisk.”Hárbarðr kvað: “Ek mynda þér þá trúa, nema þú mik í tryggð véltir.” (sts. 32–34)(Hárbarðr said: “Your help I'd have needed then, Þórr, to hold the linen-white girl.” Þórr said: “I'd have helped you with that, if I could have.” Hárbarðr said: “I'd have trusted you then, had you not betrayed my trust.”)In his second utterance here, Hárbarðr closely adopts Þórr's syntax (“Ek mynda þér þá . . . ”), which itself reproduces the conditional mood of Hárbarðr's first statement. Clover describes these lines starkly as “interstitial verbal byplay” (1979, 135). Bax and Padmos see the exchange as continuing to follow flyting conventions in an unremarkable way: Hárbarðr “jestingly . . . admits that the help of Þórr would have been useful then”; Þórr, “trying to top his opponent, continues the joke by stating a precondition of his helpfulness (being present at the moment),” thus providing a “pseudo-excuse”; Hárbarðr then reacts to this “clever stroke” with a “senna-like insult” (Bax and Padmos 1983, 163). Arnold acknowledges, however, that it is difficult to work out what precisely is being communicated here, and contemplates whether Óðinn's mask is slipping slightly, such that the audience witnesses a “moment of father-son intimacy and verbal negotiation,” quickly punctured by an “unexplained cause for recrimination” (2014, 14). Given that the tone initially seems jocular, Arnold suggests that if a framework from pragmatics were to be utilized to interpret these lines, H. P. Grice's Co-operative Principle would be more appropriate than that of a Face Threatening Act (Arnold 2014, 14; Grice 1975).These lines demonstrate the capacity of repetitious dialogic syntax to frame other kinds of utterance, in addition to and alongside competitive speech acts, which are not straightforwardly hostile, competitive, or intimidatory. The third stanza (st. 34) has layers beyond simple antagonism. Although Hárbarðr's echo does not turn a full or partial echo into a question, thus differing from the echo responses described by Channon, Foulkes, and Walker (2018), his utterance can nonetheless be understood as a different kind of “repair indicator,” in that it signals a latent absurdity underlying Þórr's offer of hypothetical help: Þórr cannot be relied upon, and this has become clear, Óðinn asserts, alluding to an unspecified betrayal. Similarly, the ability of echo to signal “a degree of thematic agency, or independent interest over [a] proposition” in answers to polar questions (as described by Enfield et al. 2019, 292) may also be of relevance here. Óðinn uses close lexical and syntactic parallelism to center his experience of Þórr's help, or lack thereof, asserting an independent interest over the idea, born of bitter experience. It has been observed that traditions of verbal dueling often rely on fictitious claims (Parks 1990, 114), a convention observable in Hárbarðsljóð (Bax and Padmos 1983, 154–5, on sts. 4–5, particularly), but this exchange shows the poem's echoing replies to also be capable of shading into hypothetical, wished-for, or conditional claims, imagining different possible pasts and futures. A more wide-ranging concept of verbal echo as a device that can comfortably facilitate different kinds of stance is therefore helpful here. In a statement from which inferences of reprimand, disappointment, anger, and melancholy could all be drawn, Óðinn directs this moment of possible intimacy towards a re-assertion of the interpersonal distance between himself and Þórr.Compared with Þórr, Hárbarðr has often been credited as the more skillful appropriator of his opponent's speech, particularly when he seemingly ironically borrows the formula that Þórr uses to introduce his martial exploits (“Ek var austr,” sts. 23.1, 29.1) in order to introduce an account of sexual exploits (st. 30.1) (Clover 1979, 137).20 Nonetheless, Þórr does also take up echo as a mode of retort, as when the pair discusses where Hárbarðr got his “hnœfiligu orð” (st. 43.2) [taunting words]: Hárbarðr kvað: “Nam ek at mǫnnom þeim inum aldrœnum er búa í heimis skógum.”Þórr kvað: “Þó gefr þú gott nafn dysjum, er þú kallar þat heimis skóga.” (sts. 44–45)(Hárbarðr said: “I got them from those ancient men, who live in the woods of home.” Þórr said: “You give a good name to burial mounds, when you call them the woods of home.”)Hárbarðr's reference to “the woods at home” is fairly obscure: Orchard (2011, 291) believes it must have been a kenning for burial mounds, given that Óðinn is known to gain knowledge from the dead; Larrington (2014, 273) also concludes that the “ancient men” are presumably the dead. The reference to “home” may nonetheless be appreciated as another suggestion of potential intimacy and familiarity, again quickly punctured. This time, it is Þórr who scathingly reasserts the interpersonal distance, degrading the source of Hárbarðr's words and simultaneously asserting his own understanding of the status and significance of the source they are discussing.Elsewhere, replicated features of speech enable other ambiguous reflections on the mysterious status of the pair's relationship and their current disagreement. This is especially noticeable in the early stages of the poem when Þórr calls attention to the uncertain nature of the “dispute” itself: Ferjukarlinn kvað: “Hárbarðr ek heiti, hylk um nafn sjaldan.”Þórr kvað: “Hvat skaltu of nafn hylja, nema þú sakar eigir?” (sts. 10–11)(The ferryman said: “Hárbarðr I'm called, I seldom hide my name.” Þórr said: “Why should you hide your name, unless you have some disputes?”)This question can be interpreted as an aggressive piece of conjecture, assuming the ferryman's quarrelsome intentions, or (more in keeping with what scholars such as Clover have interpreted as Þórr's bewildered attitude) it can be taken at face value, as a genuine question. Hárbarðr responds with another mirroring reply, which, like Þórr's, adopts the second clause of the preceding statement: “En þótt ek sakar eiga . . . ” (st. 12.1) [Whether or not I have a dispute . . . ]. Hárbarðr then announces he would nonetheless defend himself from people like Þórr (st. 12): a pejorative stance is clear at this stage. Later, however, Þórr's initial comment about the sǫk (dispute) is echoed in a less straightforwardly abusive context. This echo is triggered as Þórr threatens Hárbarðr for mentioning Skrýmir's glove (st. 26): Þórr kvað: “Hárbarðr inn ragi! ek mynda þik í hel drepa, ef ek mætta seilask um sund.”Hárbarðr kvað: “Hvat skyldir þú um sund seilask, er sakir ro alls øngar?” (sts. 27–28.2)(Þórr said: “Hárbarðr, you deviant! I'd beat you into Hel, if I could reach across the strait.” Hárbarðr said: “Why should you reach across the strait, since there are no disputes at all?”)On one level, Hárbarðr wrong-foots Þórr with this playful reply, denying the by-now evident disagreement. He creates another hinge-like structure by folding elements of the last part of Þórr's utterance into the beginning of his own, while calling back to Þórr's questioning syntax from seventeen stanzas earlier. He thereby asserts his verbal skill. However, although irony and competitive adroitness are both clear, the implications of Hárbarðr's utterance do not necessarily end here. The strange status of the disagreement itself is highlighted obliquely, just as when Hárbarðr asserts Þórr's prior betrayal of his trust. The echo implicitly poses a question with implications reaching beyond the assertion of verbal skill: what is the real interpersonal problem, or set of problems, at the heart of this altercation?The actual causes