Walter the Reviser
The above certainly does not document every change that Walter makes as he revises. It provides enough of an overview to address one major issue concerning the textual state of the De nugis curialium. The minor revisions all point to the same direction of change: if we grant Walter even the least bit of authorial competence, in each of the alterations discussed above the stories found in distinctiones 4 and 5 seems to be the earlier version of their counterparts in distinctiones 1 and 2. The major revisions, which focus mainly on narrative structure, also point to this direction of revision. However, to my mind the minor changes most convincingly demonstrate that the doublets do indeed result from the process of revision. Otherwise, one would have to suppose that as Walter rewrites, he is reducing his alliteration, choosing less striking or appropriate words, and generally impairing the rhetorical success of his work. A few may prefer this view of Walter; I find it unlikely. The man who could gleefully write “si me ruditus ruditas ridiculum reddiderit” almost certainly did not restrain his alliteration and wordplay as he revised.105 Alternatively, one could suppose that the doublets are merely different versions of the same story, recorded decades apart. However, the direct verbal similarities, often exact, in the doublets prove that Walter had an earlier version in front of him as he reworked his prose. The unfinished textual state of the De nugis curialium has even tidily provided us with another group of doublets with which we can compare Walter’s revisions. In distinctio 1, Walter twice recounts tales of the Carthusians and the Order of Grandmont. And while these doublets revisit the same material, they do not represent the same story at different stages of revision; Walter has composed them at different times with different aims. Aside from the broadest of generalizations, they share no direct verbal or thematic likenesses. After comparing these tales, the revised nature of the other doublets stands out in stark contrast.
Comparing Walter’s revision of his own tales with his use of other sources also demonstrates that Walter was a careful reviser of his own work as well as that of others. Like most medieval writers, Walter saw no harm in reworking stories from other authors and sources, but in this respect, Walter is no verbatim transcriber. When he uses another source, he tends to shape it to his needs, editing and rewriting it with a strong focus, unafraid to make radical, and felicitous, changes. Nowhere in his work do we see the wholesale borrowing that is not altogether uncommon in medieval composition, as in the Gemma ecclesiastica of Gerald of Wales, which lifts several passages from Peter the Chanter.106 For example, Walter has carefully reworked a passage ultimately from Cicero’s De officiis to good effect in his story of Earl Godwine of Wessex.107 The tale of Sadius and Galo, one of Walter’s most celebrated pieces, has been skillfully stitched together from several narrative sources.108 And Walter’s refashioning of a story from the collection known as the Analecta Dublinensia is, according to one critic, “more coherent and more satisfying” than the original.109 While an examination of Walter’s use of sources is outside the scope of this book, initial studies all agree that Walter was an adroit adapter who carefully modified earlier material. Clearly, Walter took the same approach when it came time to revise his own work.
That Walter revised the material in the De nugis curialium sheds light on another textual mystery. The Dissuasio Valerii, Walter’s anti-matrimonial tract found in distinctio 4, circulated separately from the De nugis curialium. Its popularity—it remains most famous as one of the major sources for Jankyn’s “book of wikked wyves” in The Canterbury Tales—has ensured a complex and rich manuscript tradition. The work’s most recent editors have argued that the transmission of the text was bifid, meaning that all witnesses of the Dissuasio stem from two families representing separate, irreconcilable archetypes.110 These archetypes, which they term “alpha” and “beta,” may well represent two different authorial versions of the Dissuasio Valerii.111 Alpha appears to be the earliest, and seems to have been in circulation by 1184.112 The copy of the Dissuasio found in the De nugis curialium, on the other hand, belongs to the beta family of manuscripts (though it is interestingly not the best representative of this family). This state of affairs would to be expected if the beta family represents a revised version of the Dissuasio, since by Walter’s own testimony the Dissuasio was already popular by the time Walter revisited it in the De nugis curialium. It seems then, that these two families of manuscript are best thought of as earlier (alpha) and revised (beta) versions of the Dissuasio Valerii. This works well with what we know about the circulation of the alpha and beta groups. The alpha group appears to have spread rapidly throughout the continent and is by far the larger of the two families. Only eight manuscripts belong to the beta group. It is very revealing that in addition to the copy of the Dissuasio in the De nugis curialium, only one other manuscript, British Library Additional 34749, names Walter Map as the text’s author—and both of these belong to the beta (revised) group.113 In other words, the only manuscripts that name Walter Map as the author of the Dissuasio Valerii belong to the small family of manuscripts that I believe represents the revised version of the text, a situation that accords with what Walter himself tells us: that he only added his name after the Dissuasio became incredibly popular. Tellingly, lines 284–317, which contain an overtly Christian exhortation and thus destroy the illusion that the text is indeed ancient, seem to have their origin in the beta version.114 Walter seems to have added this passage only after he revealed his own authorship of the Dissuasio. Only a thorough comparison of the alpha and beta traditions can confirm that he did in fact produce two separate versions—a task that the lack of a good alpha edition renders impossible for the moment. Nonetheless, internal evidence from the De nugis curialium has already shown Walter to be a thorough reviser. It is not surprising in the least that he revised his most successful work.
As a reviser, Walter is no anomaly in the twelfth century. Gerald of Wales wrote five versions of Topographia Hibernica.115 Peter of Blois seemingly could not resist the urge to fiddle with his letters, and later in his life he gave his letter collection a fairly radical overhaul.116 Likewise, the textual tradition of Nigel Wireker’s Speculum stultorum suggests that this popular satire also underwent one revision.117 Walter, like these and other authors, worked through his text with care, tinkering with words, rearranging phrases, and repurposing stories. Importantly, this care does not suggest a careless anecdotist who has hastily jotted down witty sayings. Walter Map may in fact have more in common with William Langland, the most well-known reviser in medieval England; it is a pleasant coincidence that an important copy of Piers Plowman resides in Bodley 851, a happy companion to the only copy of the De nugis curialium.
Chapter 3
Glosses and a Contrived Book