The Dream Women of Gísla saga

Abstract

MODERN SCHOLARSHIP on Gisla saga Surssonar has devoted a considerable amount of discussion to the women of Gisli's dreams. The precise function and nature of these women, however, have been somewhat obfuscated by their dual representation, namely in verse and in prose. Despite striking incongruities between these alternate accounts, modern scholars have tended to take the compiler's reading of the poetry at face value. When the dream stanzas are instead analyzed in isolation from the prose and with attention to their respective moods and religious functions, the juxtaposition of the two women suggests a dualistic afterlife strongly reminiscent of medieval Christian visionary literature. The images by which this afterlife and its proximity are heralded, however, are traditional northwestern European motifs. Thus a blend of Christian and vernacular elements appears to serve two motifs of warning: the warning of death follows a vernacular tradition while the didactic message of what follows is presented in a Christian duality. PROSE AND VERSE Although the various stanzas of a saga need not have followed the same route, Gisla saga (1) is a good example of the classical paradigm in which the compiler builds his prose narrative around preexisting verse. Since the poetry was thus typically not his own, its original intent may have often been unknown to him. Accordingly, the poetry in Gisla saga at times falls to correspond to the accompanying prose, signaling an imperfect merger. An instance of this incompatibility can be found in stanza 22: Sagt hefk enn fra drum oddflaums vidum draumi, Eir, vardat mer, aura, ordfatt, es munk lata. Verr hafa vapna snerru vekjendr, peirs mik sekdu, brynju hatrs ens bitra beidendr, ef nu reidumk. (77) (Again have I told [warriors] of my dream, [woman], of when I will die; I did not run short of words. The [warriors] who outlawed me will have the worse of it if they make me angry now.) (2) Despite the specific references in this verse, it is spoken in a virtual narrative vacuum. Both the beginning of the chapter and the prose following the stanza are characterized by the passing of time (pegar er varar (75) [as soon as it became spring]; eru nu kyrr tidendi (77) [things were now quiet]). The prose in between these temporal markers has Gisli move back to Geirpjofsfjordr, and it speaks of frequent dreams dominated by the evil dream woman. Gisli then recites two stanzas on the contents of his dreams, followed by stanza 22 quoted above. This verse, however, relates to dreams only inasmuch as the poet claims to have recounted his dream. Not only is there no reference to the content of the dream in question, but Gisli has never recounted any dreams of this group to oddflaums vidir, that is to warriors or men. The only comparable situation is when he tells his dream concerning Vesteinn's murder to Porkell, conceivably in the presence of others (chapter 14-). (3) That dream, however, survives only in the prose. The dreams involving female characters are told to Audr alone. Stanza 22 also refers to the poet's outlawry, news of which in the prose narration had reached Gisli three chapters earlier (chapter 21). There is, thus, no reason for him to bring it up again at this point in the narrative. It appears that a seam of the saga's structure is here visible, suggesting that the poetry and prose were not composed by a single author. More such inconsistencies can be found: O'Donoghue, for instance, has remarked on the discrepancies between prose and poetry following stanza 31, and similar considerations apply to at least one of the dream stanzas (see below, 12-4). It may thus safely be assumed that much of the poetry of Gisla saga predates the prose. (4) When dealing with such a composite work, taking the saga as a unified whole is potentially misleading. This danger has found expression in O'Donoghue's acute observation that the dream poetry of Gisla saga never distinguishes explicitly between a good and an evil dream woman (162-3). …

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 5 citations exist upstream.