The Beowulf Poet's extraordinary facility in using a vast and diverse word-hoard has long excited students of the poem. Among the critical studies, discussions of vocabulary rank high in number, and almost every conceivable approach to the subject has been investigated either in part or with a high degree of thoroughness. Single words, such as ealuscerwen, and groups of related words, such as rime-words, kennings, and words of Christian content or reference, have received close attention, as well as larger lexical patterns, such as variation and the formulaic texture, while further studies have compared the vocabulary with that of other Old English poems or Nordic literatures. Aside from purely lexicographical or etymological inquiries, there are three perspectives to which these many discussions generally belong: 1) descriptive: usually statistical observations about the number of compounds relative to simplices or of formulas relative to the whole vocabulary of the poem, or a comparison of the frequency of certain lexical types with other poems, or a classification of the habits of word-formation; 2) figurative and appellative: the types of verbal figures and their analogues elsewhere in Old English and Old Norse; and 3) usage: the use of words in particular contexts or for specific effects, and the structural use of synonymic substitution and variation. The first emphasis is important because it reveals the composition and its formative strata of the poem's total vocabulary, and also the lexical relationships with other poetry or poetic traditions. The second serves to isolate a lexical stratum which is by nature exclusively poetic and to observe how much of this stratum is probably original and how much traditional. But it is the third perspective which is interested most essentially in the poet, since here the attempt is made to discern the many ways by which he has used language significantly to dramatize, emphasize, elucidate, intimate, and so on. Much that has been written in this category has concerned itself with the larger patterns of variation as a characterizing, describing, or structural device, rather than with smaller, more confined, strokes of verbal association and verbal play. A well-known instance of the latter is the epithet for Grendel, healoegn (142) which, in its context, wherein a bona fide hall-thane anxiously seeks out a hiding place as protection against the intruder, may with complete justification be termed ironic, and the same thing may be said of a similar appellation used later for both Grendel and Beowulf, renweardas (770), There are also hints here and there that the poet may have been influenced by learned Latin figures. Many years ago Albert Cook compared flod blode weol (1422; Exodus 463, flod blod gewod) to Aldhelm's fluenta cruenta (De Virginitate, 2600), and more recently H. D. Meritt called attention to the similarity between Hrothgar's warning that in death “eagena bearhtm / forsiteo ond forsworces” (1766b-67a) and Aldhelm's “ferreus leti somnus palpebrarum conuolatus non tricaverit” (De Virg. Prose, 321.7, ed. Ewald). It is in the smaller strokes, I think, that the poet's acumen and craft are most incisively contained, and it is to some of these that the present discussion is devoted.
Summarizing the solution space in tumor phylogeny inference by multiple consensus trees