Philology at the beginning of the twenty-first century
The distinguished paleographer, the late Malcolm Parkes, was accustomed to opine in conversation that “the greatest mistake a paleographer makes is to forget the nature of the text being copied.” The axiom is a powerful one, with relevance not simply for the subdiscipline of paleography but also for the wider philological enterprise of which paleography is part. In this chapter, part of a much larger ongoing research program on the afterlives of medieval texts, a set of texts from the period under review that survive in more than one version will be examined. It will be demonstrated how certain characteristics of these texts – spelling, punctuation, certain paleographical/bibliographical characteristics and their layout, all broadly speaking philological – can be related intimately to their textual function. The wider theoretical framework for the chapter, therefore, may be characterized in broad terms as philological and pragmatic.
The term philology has of course a wide range of meanings, and indeed these meanings have changed through time. The birth of the “new philology,” arguably the nearest thing to a real (as opposed to a claimed) paradigm shift that has ever happened in linguistic enquiry, is traditionally dated to a single event: Sir William Jones’s Third Anniversary Discourse to the Asiatic Society of Bengal, delivered in 1786. Sir William’s speech on that occasion included the following famous passage:
The Sanskrit language, whatever may be its antiquity, is of a wonderful structure; more perfect than the Greek, more copious than the Latin, and more exquisitely refined than either, yet bearing to both of them a stronger affinity, both in the roots of verbs and in the forms of grammar, than could possibly have been produced by accident; so strong indeed that no philologer could examine all three, without believing them to have sprung from some common source, which perhaps no longer exists; there is a similar reason, though not quite so forcible, for supposing that both the Gothic and the Celtic, though blended with a very different idiom, had the same origin with the Sanskrit; and the old Persian might be added to the same family, if this were the place for discussing any question concerning the antiquities of Persia.1
Although its novelty has been questioned,2 Jones’s statement remains – once we look past its eighteenth-century terminology (e.g., copious, i.e., “elaborated”) – a concise outline of the comparative method, whereby languages are compared in order to reconstruct the nature of their common ancestor. This approach was “the new philology.”
The new philology expressed itself more comprehensively through the tree diagrams of Jacob Grimm (1785‐1863), enunciator of the eponymous Law, which were used to reconstruct the archetypal language of common ancestor-languages through the analysis of extant cognates. But it also expressed itself in the study of textual relationships, something hinted at by Jones’s reference to antiquities, through the textual criticism of Grimm’s contemporary Karl Lachmann (1793‐1851). Lachmann’s stemma codicum – or family tree – of manuscript witnesses, used to reconstruct the archetypal text and thence the author’s original conception of the work, is clearly related to the tree models of languages developed by Grimm and his successors.
Grimm and Lachmann underpin pretty well all the philological enterprises of the nineteenth and twentieth centuries, from the great Neogrammarians to what became the Oxford English Dictionary, and philological “rigor” made subjects new to nineteenth-century universities – such as the discipline that later became “English studies” – respectable. The philological tradition continued – indeed it underpinned the disciplinary formation of numerous Anglicists until quite recently – and it never lost the range of concerns that Jones enunciated: a combination of pattern-seeking beside an intense empirical focus on texts (Jones’s “antiquities”). But, by the final decades of the twentieth century, philology seemed to have run its course and was ripe for rethinking. Its prominence in linguistic inquiry had been overtaken in many circles by the rise of linguistics as a discipline distinct from textual study; generative linguistics, the dominant model in the United States particularly, was in its focus on formalism much more akin to philosophy or mathematics. And as an approach to text, philology seemed under-theorized in comparison with, say post-modernism.3
However, the influential “new philologists” (really, it may be argued, the “‘new’ new philologists”) who contributed to Stephen Nichols’s special number of Speculum in 1990 drew on a kind of critical approach that found its expression, for medievalists at least, in the writings of Derek Pearsall and others from the late 1970s onwards.4 Inspired by current theoretical trends in the humanities, the “new” new philologists wanted a more theoretically sensitive approach to textual study which took on board post-modern, destabilizing thinking; and they undertook this task by reversing the telescope, as it were, setting aside the focus on archetypes and concentrating on variance – Bernard Cerquiglini famously wrote “in praise of the variant” – and what Paul Zumthor had some years before called mouvance.5
The “new” new philology differed from the older variety in not emphasizing the linguistic aspect of the enterprise, but in recent years it has, in some circles at least, mutated into something else, which brings it back into engagement with its linguistic side: historical pragmatics. Pragmatics for linguists is the study of how language works in situations, and is increasingly interesting for linguists working with cognitive models of language: “a shift seems to be taking place in linguistics towards pragmatic approaches … with context playing a more prominent role than before.”6 Historical pragmatics, as its name suggests, applies this insight to the past, and is necessarily focused on written texts. And although much research in the field has hitherto been devoted to the discussion of more obviously “linguistic” phenomena such as the grammatical/lexical expression of (im)politeness,7 other developments of the discipline have taken on board quite delicate “textual trace” features such as punctuation.8
The rest of this chapter will focus on a number of texts that, originating in the Old and Middle English periods, demonstrate processes of mouvance in quite delicate ways. I will conclude by returning to some of the implications of the approach offered here.
Laʒamon’s Brut
The language of London, British Library, MS Cotton Caligula A.ix (Part I), dated to the second quarter of the thirteenth century, has been localized to the south-west Midlands, more specifically north-west Worcestershire.9 The manuscript contains one of two surviving texts of the early Middle English epic/romance poem Laʒamon’s Brut, and the Caligula text is generally considered to be closer to the authorial original. Passage 1 below is based on Brook and Leslie’s edition of the opening lines of the poem, but has been corrected against the manuscript. The first initial <A> is historiated, containing the famous drawing of Laʒamon bent over his book (the only such elaborate initial in the whole of the Caligula text).10 While in verse, the passage is written out (as was the case with Anglo-Saxon poetry), in the same manner as prose, although (unlike Anglo-Saxon verse) in double columns on each page. I have marked the end of manuscript lines with |, while flagging the verse structure in lineation.11
Passage 1
(There was a priest among the people who was called Laʒamon; he was the son of Liefnoth, may the Lord be merciful to him. He dwelt at Areley, at a noble church, upon the banks of the River Severn, where it seemed splendid to him, right beside Radestone; there he recited his Missal. There came into his mind a most splendid idea, that he would tell concerning the English of the most outstanding men: what they were called and from where they came, who first possessed the land of the English, after the flood, that came from God, that killed everything alive that it found there, except for Noah and Shem, Japhet and Ham and their four wives who were with them in the ark. Laʒamon travelled widely throughout this land, and secured the noble book which he took as an exemplary narrative; he took the English book which Saint Bede had created, a second he took in Latin made by Saint Albin and the noble Augustine, who brought baptism here. The third book he took, that he placed in the middle, was made by a French scholar who was called Wace, who well knew how to write; and he gave it to the noble Eleanor, who was queen of the noble king Henry.)13
The verse form of the text is based around the half-line unit characteristic of Anglo-Saxon poetry, although the syntax of the passage is “looser,” with considerable use of prepositions, required given the obscuration of the comparatively complex Old English inflectional system. It has been argued that the deployment of short units in sequence, e.g., line 19, is more characteristic of the so-called “popular” style of late Old English verse than that of poetry traditionally seen as more archaic, e.g., Beowulf.14 Litterae notabiliores (i.e., capital letters) are deployed in general to mark steps in the argument, although they are also sporadically used for personal names. However, in general the punctuation of the passage is closely tied to its verse structure, with symbols known as the positurae, in origins used for “pointing” liturgical texts, deployed to assist declamation. Thus the mid-line caesura is generally flagged by the symbol known as the punctus elevatus, viz. ؛, used to indicate a major medial pause, and the end of lines by the simple point or punctus; it will be observed that, in this passage at least, the occurrence of the caesura corresponds to a break between periodic units. In sum, the punctuation is designed to assist oral delivery, while the “loose” syntactic structure, as was commonplace well into the early modern period and indeed beyond, reflects “speech-like” usage.
As flagged above, Laʒamon’s Brut survives in two versions. The second version, London, British Library, MS Cotton Otho C.xiii, dated from the second quarter of the thirteenth century with language localized by LAEME to north-west Wiltshire, has generally received less attention than the Caligula text. Not only was the manuscript severely damaged in a disastrous fire at Ashburnham House in 1731, but also the Otho text was clearly always an abbreviated form of the poem. For that reason, the Otho manuscript has never received the same kind of attention as the Caligula text.
The opening of the Otho Brut was lost in the Ashburnham fire, but not before it had been transcribed by the distinguished paleographer and librarian Humfrey Wanley (1672–1726). Wanley’s transcription has been the main source for all modern editions of these lines, such as Brook and Leslie. Brook and Leslie present the Otho opening as follows:
Passage 2
This edited version (passage 2) allows us to compare the vocabulary of the two versions of the text, and it is at once clear that the Otho text represents a “modernized” version of the Brut, replacing (e.g.) the Old English forms seemingly characteristic of Laʒamon’s “antiquarian sentiments.”16 Thus archaic (or archaistic) forms such as Caligula’s drihten (line 10) or leoden (line 1) appear as God (cf. the confused form Driste, line 2) and londe in Otho. (It also demonstrates, inter alia, what were probably Wanley’s interventions; the only mark of punctuation deployed is the punctus, whereas we know from elsewhere in the Otho manuscript that the punctus elevatus must also have been used.) Such regular departures from what must have been “difficult readings” have added to the general view that the Caligula manuscript is to be preferred for study as being closer to the author’s original conception of the work.
However, more recently scholars have been more willing to celebrate Otho’s virtues. Jane Roberts, commenting on the Otho scribe’s calligraphic qualities, suggests that “this must once have been rather a pretty book,”17 while Elizabeth Bryan, most notably in an important monograph from 1999, has drawn attention to several features of the Otho manuscript that are worthy of special attention.18 One such feature, discussed by Bryan in detail and very relevant to the themes of this chapter, is the deployment of a comparatively sophisticated set of punctuation marks, litterae notabiliores and paratextual features.
To demonstrate the deployment of such features, we might examine the following transcribed passage from later in the Otho version of the poem. The modern lineation is that of Brook and Leslie, but again I have shown the layout of the manuscript using | to indicate line endings. Lacunae/damaged letters in the manuscript are indicated by an asterisk (*), each occurrence flagging a presumed missing letter. Bold or engrossed letters are used to indicate decorated litterae notabiliores.19
Passage 3
Here by contrast is a diplomatic transcription of the Caligula text of the same passage, exactly as it appears in the manuscript (e.g., including the intrusive <g> in line 1958):
Passage 4
(And at that same time a marvellous sign came here, that never before nor afterwards came hither; there came here from heaven a miraculous flood; for three days it rained blood, three days and three nights [it was a very great misery]. When the rain had gone another sign came here swiftly; black flies came here, and floated in men’s eyes, in their mouths, in their noses, causing them to lose their very lives. Such a swarm of flies was here that they ate the corn and the grass; all the people who lived in the land were wretched. After this came such a great plague that few were left alive. Afterwards here came a severe event, that King Riwald was dead. King Riwald had a son who was called Gurgustius; he governed the land for half a year, and then he fell dead. After then came Sisilius; he was dead here at once. Then came Lago, who lived for eight weeks; then came King Mark, who was king for thirty weeks. Then came Gorbodiago; he was a good king for five years. The king had two sons, both accursed. The older was called Fereus; the younger was called Poreus. These brothers were so mad, and so contrary.)
It will be observed immediately that the Otho text has a more sophisticated pattern of punctuation than Caligula, in the sense that the decorated initials stand out on the page. As Bryan points out, although there is a fairly close link between the textual organization of Caligula and Otho in the deployment of punctuation marks (including, sporadically, litterae notabiliores), in the Caligula manuscript “the paragraph instructions do not stand out to the eye, especially by comparison to the rubricated names in the margin of that manuscript. In [Otho], however, the eye is drawn to the large colored initials first.” She goes on to argue as follows:
By distinguishing between instructions for large ornate initials and less eye-catching elements, [Otho’s] scribe transmitted or created a hierarchy of elements that does not exist in [Caligula], even where [Caligula’s] placements match [Otho’s]. [Otho’s] hierarchy is systematic. It gives precedence to regnal succession as the structuring principle of the work, it focuses attention on the Arthurian section through increased density of initials, and it devalues most narrative sections that describe British wars with Rome, including Arthur’s.22
These two versions of the text, although dating from roughly the same time, are therefore very different in approach, and this difference manifests itself pragmatically, in terms of the ways in which the texts have been presented for the reader’s use. The Caligula text is clearly designed for readers who can find their way round the text with comparatively minimal direction and can be expected to collaborate with the scribe in the interpretation of the text, assisted by their own established habits of vernacular reading. Such readers would have no difficulty with Laʒamon’s use of terms such as drihten or leoden. By contrast, the Otho text is more directive, drawing the reader’s attention to a particular imagining of history. It could be argued that contemporary readers received the Otho text, rather than collaborated in its interpretation.
Two versions of Ancrene Riwle
Ancrene Wisse (“A Guide for Anchorites”) or Ancrene Riwle (the work has two titles) was composed at approximately the same time as Laʒamon’s Brut, and seems to be part of the same cultural milieu. Traditionally Ancrene Wisse has been seen as the “end of the line” for a particular kind of vernacular prose composed in the English south-west Midlands: a last gasp of the prose tradition that had included Wulfstan and Ælfric. Ælfric’s homilies were, however, still being copied, modified, and read at around the time the Ancrene Wisse author was composing his work.23
Ancrene Wisse/Riwle had a considerable cultural impact in its area, with nine English manuscripts or distinct manuscript fragments, two translations into French, and one translation into Latin. Most date from the century after its composition, c. 1200‐20, but one in particular dates from much later: the Vernon manuscript miscellany in Oxford, Bodleian Library, MS Eng. Poet. a.1, dating from 1390‐1400.
The Vernon manuscript has received a lot of attention recently, most notably from Wendy Scase and her associates working on the West Midlands Manuscript Project, and I myself, in a volume edited by Scase, have written on some of the linguistic features of the texts in the manuscript and how they can offer us insights into various cultural developments.24 The manuscript – a vast object, weighing some 22 kilos – seems to have been a kind of “millennium ark,” a repository of 370 poetry and prose texts on devotional or moral themes.
The Vernon text of Ancrene Riwle is an attempt to reproduce a text that originated some two centuries before the creation of the Vernon Manuscript. It is therefore unsurprising that the Vernon scribe found it very challenging to turn this archaic text into something more readable for his contemporaries, and an analysis of the changes which appear in the Vernon text when compared with the other early Middle English versions of the work is very illuminating for the purposes of this discussion.
The Vernon text of Ancrene Riwle is most closely related in stemmatic terms to the Nero text of the work, i.e., London, British Library, MS Cotton Nero A.xiv, which dates from the middle of the thirteenth century. The Vernon and Nero texts seem to derive from a common lost ancestor in the stemma codicum suggested by Eric Dobson.25 There is also some evidence that the text as it survives in the Vernon Manuscript has been influenced in some way by the most authoritative version of the work, Cambridge, Corpus Christi College, MS 402, i.e., there has been contamination of the Vernon/Nero tradition by the Corpus tradition, not only in substantive terms (as pointed out by Dobson) but even in terms of layout.
As with the versions of the Brut, linguistic comparison is also very illuminating. I have commented elsewhere on the various features that have been changed, e.g., in terms of spelling (reflecting sound changes), and also in terms of vocabulary and grammar.26 So, for instance, the word ferde (army; cf. Old English fyrd) is at one point replaced by the less metaphorical, but also less archaic, form strengþe, and at another point by host. Archaic dole (host) is replaced by Book; archaic onont (with respect to) has been replaced by more commonplace on; ʒet (sends forth) is replaced by ʒeldeþ (yields); ereste (most original) has been replaced by Furste; i efned (likened) has been replaced by I.likne, licamliche (bodily) by bodiliche. The native form foes is at one point replaced by enemyes – though fo is later retained – and bitechen (bestow) is replaced by bi taken (given). There has therefore been some linguistic updating between Nero and Vernon.
But one area that has not been traditionally considered as part of “linguistic” discourse – though it is an argument of this chapter that it should be – is punctuation. Here are parallel versions of the opening of Book V, from the Nero and Vernon texts respectively, derived from the diplomatic editions published by the Early English Text Society:
Passage 5
Monie kunnes fondunge beoð ine þisse uorme dole. and misliche urouren. & moniuolde saluen. vre louerd ʒiue ou grace ðet heo moten ou helpen. of alle þeo oðre. þeonne is schrift. ðe biheueste. of hire schal beon þe vifte dole ase ich bihet þeruppe. and nimeð ʒeme hu euerich dole ؛ ualleð into oðer. ase ich er seide. her biginneð ðe uifte dole of schrifte.
Of two þinges nimeð ʒeme of schrifte ؛ iðe biginnunge. þet forme þing. of hwuche mihte hit beo. þet oðer þing. hwuch hit schulle beon. þis beoð nu. ase two limes. and eiðer is to dealed. þe uorme؛ o six stucchenes. ðe oðer ؛ o sixtene. nu is þis of ðe uorme.
Schrift haueð monie mihtes. auh nullich of alle ؛ siggen buten sixe. þreo aʒean ðe deouel. & þreo onont us suluen. schrift schent þene deouel. ┐ hackeð of his heaued. ┐ todreaueð his ferde. schrift wascheð us of alle ure fulðen. ┐ ʒet us alle ure luren. ┐ makeð us godes children. and eiðer haueð his þreo. preoue we nu alle.27
(There are many kinds of temptation in this preceding part, and many comforts, and many and various remedies. May our Lord give you grace that these may help you. Of all the others, it is confession that is the most useful, concerning which the fifth part must be, as I promised above. And take heed how each part leads into the next, as I said before. Here begins the fifth part concerning confession.
Pay attention to two things concerning confession. In the beginning, the first thing, of what power it is; the second thing, what it must be. These now are like two branches, and each is to be divided; the first into six sections, the second into sixteen. Now for the first.
Confession has many powers, but I do not wish to speak of them all, [but will] speak of only six. Three (are) against the devil, and three with respect to ourselves. Confession confounds the devil, and cuts off his head, and routs his army. Confession washes us of all our filth, and returns to us all our losses, makes us God’s children. And each has his three [parts]. Let us now demonstrate everything.)
Passage 6
Mony cunne fondynges. is I. þis feorþe Bok. Moni diuerse sunnus. & moni maner saluen. Vr lord ʒiue ou grace þat heo ow moten helpen. Of alle þe oþure þenne؛ is schrift þe beste. Of hire schal ben þe fyfþe Bok. as ich bi heet þervppe. And nymeþ ʒeme how vch a Bok. falleþ into oþur؛ as ich er seide.
Her beginneþ þe fyfþe Book.
TWo þinges nymeþ ʒeme. of schrift. I.þe biginnynge. þe Furste؛ of whuch miht hit beo. ¶ þat oþur؛ whuch hit schule ben. ¶ þeos beoþ. as two limen. And eiþer is to delet ¶ þe Furste. on sixe. ¶ þat oþur؛ on sixtene parties ¶ Nou is þis؛ of þe furste.
Schrift haþ mony mihtes. Ac I.nulle of alle؛ sigge bote sixe. ¶ þreo a ʒeyn þe deuel؛ and þreo on vs seluen. Schrift schent þe deuel. Hakkeþ of his heued. And al to dreueþ his strengþe. ¶ Schrift wasscheþ us؛ of alle vr fulþen. ¶ ʒeldeþ us. alle ur leoren. ¶ Makeþ vs. Goddes children. Eiþer haueþ his þreo. Preoue we nou alle.28
It will be clear from the transcriptions that these two texts of Ancrene Riwle are very different in appearance. As Roger Dahood has observed, the early (i.e., thirteenth-century) manuscripts of Ancrene Wisse deploy capital letters of varying sizes to indicate different levels of subdivision within the text; Dahood believes that this system was put in place in the exemplar of London, British Library, MS Cotton Cleopatra C.vi, a manuscript which, according to Dobson, was annotated by the author himself. As Dahood puts it, “Whoever first imposed the system of graduated initials was concerned that readers grasp the relationships between divisions and not just focus on discrete passages.”29 The Nero text has a similarly sophisticated deployment of litterae notabiliores.
But the Vernon text is supplied with much more thoroughgoing punctuation than is used in any of the earlier texts of Ancrene Riwle. The pilcrow or paragraph mark, viz. ¶, is used frequently throughout, varying with punctus, punctus elevatus, and punctus interrogativus. Litterae notabiliores are much more commonly employed in Vernon than in Nero, and the beginning of the fifth book is marked by an inset title (neither Corpus nor Nero mark this title); as Dahood has pointed out, the Vernon scribe “seems to have been especially concerned to make Part Five accessible for reference.”30 The comprehensive scheme of punctuation provided by Vernon, much more extensive than in Nero, is clearly designed to help the reader make sense of the text more easily.
It is fairly clear that the increased use of punctuation correlates with readers having access to a greater range of books, reading the same text less frequently, rather than reading a few books very frequently, i.e., a shift from a more intensive to a more extensive reading culture. The differences between the Ancrene Riwle in the Nero and Vernon manuscripts, therefore, correlate with some very significant cultural changes in the role of the vernacular during the course of the Middle English period. The Vernon manuscript may have been a repository, a millennium ark, but it is also something else: a precursor of the enhancement of devotional reading, particularly private, which Eamon Duffy, Helen Spencer, and others have detected in the late medieval period.31
The earliest editions of Beowulf
For the third example I will turn to the text with the longest afterlife of all those under review: the Old English epic poem Beowulf. The dating of this poem is, of course, a thorny issue – Old English poems tend to have been a long time in the making – but the sole manuscript, London, British Library, MS Cotton Vitellius A.xv, is generally dated to around the year 1000 CE. But the poem then fell into neglect, although noticed by the antiquarian Humfrey Wanley in his survey of the Cottonian manuscripts in the second volume of Hickes’s Thesaurus;32 we have already noted Wanley’s work on the Otho manuscript of Laʒamon’s Brut. Wanley described the poem as Tractatus nobilissimus Poeticè scriptus (a most noble treatise written in poetry), and provided a transcription of the opening and of a passage from a little later in the poem. These transcriptions have special value since they were undertaken before the Ashburnham House fire of 1731.
Wanley’s transcriptions are presented in Hickes’s Thesaurus in the special Anglo-Saxon font favored by many antiquarian editors until well into the nineteenth century. Although Wanley clearly recognized that the work was a poem (Poeticè), he followed the Anglo-Saxon practice of having the text presented as prose.
Passage 7
Hwæt we garde na. in gear dagum. þeod cyninga þrym gefrumon hu ða Æþelingas ellen fremedon. Oft Scyld Scefing sceaþena ðreatum monegum mægðum meodo setla ofteah egsode eorl syððan ærest wearð feasceaft funden. he wæs frofre gebad weox under wolcnum weorðmyndum þah. oð þæt him æghwylc þara ymb sittendra ofer hron rade hyran scolde gomban gyldan þæt wæs god Cyning. ðæm eafera wæs æfter cenned geong in geardum þone God sende folce to frofre. fyren ðearfe on geat þæt hie ær drugon aldor * * * are. lange hwilc him þæs lif frea wuldres wealdend worold are forgeaf. Beowulf wæs breme Blæd wide sprang Scyldes eafera scede landum in.
Here is a transcription of the same passage from the manuscript:
Passage 8
HWÆT WE GARDE|na. ingear dagum. þeod cyninga | þrym gefrunon huða æþelingas elle* | fremedon. oft scyld scefing sceaþe** | þreatum monegum mægðum meodo setla | ofteah egsode eorl syððan ærest wearð | fea sceaft funden. he þæs frofre geba* | weox under wolcnum weorð myndum þah. | oð þæt him æghwyle þara ymb sittendra | ofer hron rade hyran scolde gomban | gyldan þæt wæs god cyning. ðæm eafera wæs | æfter cenned geong ingeardum þone god | sende folce tofrofre. fyren ðearfe on | geat þæt hie ær drugon aldor * * * ase. lange | hwile him þæs lif frea wuldres wealdend | worold are for geaf. beowulf wæs breme | blæd wide sprang scyldes eafera scede | landum in.33
(Listen! We have heard of the glory of the people’s-kings, of the Spear-Danes, in ancient days, how the princes accomplished valour. Often Scyld Scefing deprived many tribes, crowds of enemies, of mead-seats. He terrified nobles after first of all being found destitute. He had comfort for this, he prospered beneath the clouds, he throve with glories, until each of those neighboring peoples over the ocean had to pay tribute; that was a good king. To that one afterwards was born a young offspring in the dwellings, whom God sent as a comfort for the people. He perceived the grievous distress that they suffered formerly, lacking a lord for a long while. For that reason, the Life-Lord, Ruler of Glory, granted fame to him. Beowulf, offspring of Scyld, was renowned; his fame sprang wide through Scandinavia.)
Comparison of the two versions indicates the odd slip – <ð> for <þ>, <þ> for <ƿ>, and <c> for <e> – and of course Wanley has imposed his own practice of capitalization. But in general Wanley’s transcription is – as one might expect from the leading Anglo-Saxonist of his age – accurate, and the odd slips are understandable, given the astonishing scale of the cataloguing that Wanley had set himself.
The poem then fell into neglect as Wanley, a busy man, turned to the task that was to dominate the rest of his life: the development of the Harleian collection. The recuperation of Beowulf had to wait until the labors of the Icelandic–Danish scholar Grimur Jonsson Thorkelin (1752‐1829), who visited Britain and Ireland in search of Danish antiquities in 1786‐1787.
The basis of what was to become Thorkelin’s editio princeps34 was the preparation of two transcriptions of the poem, now known as Thorkelin A and Thorkelin B. Thorkelin A was transcribed in an imitation Anglo-Saxon script by a professional copyist, probably James Matthews, in 1787. The transcript is rather impressively done, although there are some regular confusions, e.g., a crossed thorn was sometimes deployed in error rather than a plain one (thus curiosities such as þætah, þætara, þætone), and until quite late in the copying process Matthews regularly confused thorn and wynn (thus þeox under þolenum for weox under wolcnum). The second transcription (B), a hybrid edition/transcription, was Thorkelin’s own, written in his eighteenth-century “round hand.”35
Passage 9 (Thorkelin A)
HWÆt WEGARDE_|na ingear dagum þeod cyninga þrym gefrunon hu/ða | æþelingas ellen fremedon. oft scyld scefing sceaþen þreatum | monegum mægþum meodo setla of teah egsode eorl syððan | ærest wearð fea sceaft funden he þæs frofre gebad þeox | under þolenum þeorð myndum þætah oð þæt him æghwyle þætara | ymb sittendra ofer hron rade hyran scolde goban gyldan, þæt þæs god cyning. ðæm eafera wæs æfter cenned geong ingeardum, þætone god send folce tofrofre, fyren ðearfe on=geat þæthie ærdrug=on aldor * * * * * ase. lange hwile him þæs lif frea wuldres wealdend worold are forgeaf. beowulf wæs breme blæd wide sprang scyldes eafera scede landum in.
Passage 10 (Thorkelin B)
Hwæt We=gar De-|na ingear dagum þeod cyninga | þrym gefrunon huða æþelingas ellen | fremedon. Oft Scyld Scefing sceaþen | þreatum monegum mægþum meodo setla | of teah. egsode eorl syððan ærest weard. | fea sceaft funden, he þæs frofre gebad. | Weox under wolcnum weorð myndum þah. | Oð þe him æghwylc þara ymb sittendra | ofer hron rade hyran scolde gomban | gyldan. þæt wæs god cyning. Îæm eafera wæs | æfter cenned geong in geardum þone god | sende folce tofrofre fyren ðearfe on | geat þe hie ær drugon aldor * * * tise. lange | hwile him þæs lif frea wuldres wealdend | worold are for geaf. Beowulf wæs breme | blæd wide sprang scyldes eafera sceðe | landum in.36
After various vicissitudes, not least the alleged destruction of his notes – but not the transcripts – by Lord Nelson’s ships during the bombardment of Copenhagen in 1807, Thorkelin’s De Danorum Rebus Gestis Secul. III & IV. Poema Danicum Dialecto Anglosaxonica appeared in 1815, presented in a handsome roman font with a parallel Latin translation, a title page embellished with an emblem composed of laurel leaves, a lyre and a sword, and a rather grand dedication to Johann von Bülow, Thorkelin’s patron. Magnus Fjalldall has suggested that Thorkelin’s delay in producing his edition was in reality caused by his lack of confidence in his own editorial abilities.37 Reviews of the work can best be described as “mixed”; Sharon Turner, for instance, kindly suggested that, “As a first translation of a very difficult composition, I ascribe great merit to Dr. Thorkelin,” but he then goes on to state almost immediately that, “on collating the Doctor’s printed text with the MS., I have commonly found an inaccuracy of copying in every page.”38 Another reviewer, N. F. S. Grundtvig, was so dismissive of Thorkelin’s scholarship that one of Thorkelin’s friends, Frederik Schaldemose, as late as 1847, criticized Grundtvig as follows: “a young student who has since distinguished himself right into his old age by vulgar coarseness in his many literary quarrels, with his usual energetic mode of expression threw mud like a street urchin and loaded the old man with filth, without taking account of the many sacrifices he had performed in order to bring the old book to light.”39 But Schaldemose was fighting a losing battle, and the crushing verdict of John Kemble in his edition of 1833 now holds sway: “not five lines of Thorkelin’s edition can be found in succession, in which some gross fault either in the transcript or the translation, does not betray the editor’s utter ignorance of the Anglo-Saxon language.”40 Perhaps the most comprehensive condemnation of the edition was an unpublished initiative: the detailed and devastating collation of the Thorkelin text with the original manuscript, undertaken by John Josias Conybeare (Rawlinson Professor of Anglo-Saxon, 1808‐12) and Frederic Madden, later Keeper of Manuscripts at the British Museum.41
The authority of Thorkelin’s edition is therefore, at best, dubious, and it is to say the least unfortunate that he clearly did not understand the opening line of the poem, glossing Hwæt wegar by Latin Quomodo. In part, this was probably due to his habit – understandable in the conditions of the time – of using Matthews’s transcription rather than the manuscript itself as the basis of checking the editorial process,42 yielding for instance the reading Goban in half-line 21 below; Thorkelin’s own transcript reads (correctly) gomban. A lack of confidence also accounts for Thorkelin’s omission in his edition of the final letters in half-line 30; Matthews’s ase is accurate, whereas Thorkelin’s tise, in his transcript but omitted in his edition, makes no sense. Modern editors generally reconstruct the half-line as aldorlease.
Passage 11
For the purposes of the current discussion, however, these various ways of presenting the text are all of considerable interest, clearly relating to shifting cultural imperatives. The Old English original (Passage 8) is presented simply, designed for practiced readers who are able to use the text as a starting point for interpretation. The Anglo-Saxon scribe sees no necessity to deploy litterae notabiliores to flag names – something that was to cause problems for Thorkelin, who was to mistake inter alia the first element of the form gifstol (gift-throne) for a personal name Gif.44 Punctuation is minimal, limited to the simple punctus and to word division, the latter sometimes suggesting that our modern conceptions of Old English morphosyntax could require some revision, e.g., huða, for geaf.
The special Anglo-Saxon font used by Wanley (passage 7) was possibly one of those fonts imported from the Netherlands by John Fell, vice-chancellor of Oxford (1666‐9) and a key figure in the history of the Oxford University Press. However, it is more like the Pica Saxon developed for the seventeenth-century Germanic philologist Franciscus Junius the younger (1589‐1677), who bequeathed his books and other materials to the university.45 The font became associated with the construction of “Anglo-Saxonism” as a distinctive feature of the new, distinctively British and Protestant order that emerged after the Glorious Revolution of 1688, underpinning many aspects of contemporary antiquarianism.46 It is no coincidence that radical thinkers such as Horne Tooke and Leigh Hunt venerated the Anglo-Saxon king Alfred the Great as a legendary champion of ancient liberties. And the careful tracing of the manuscript offered by Matthews (passage 9) no doubt in addition relates to the eighteenth-century craze for facsimiles, encouraged by technical developments in lithography and aquatint.
Thorkelin’s own project, as expressed though his transcription (passage 10) and his published edition (passage 11), is similarly located in a particular historical moment. His ideological stance is clear from his preface; as Haarder and Shippey suggest,
His lengthy encomia on Hrothgar and Hygelac … are statements about the virtues of monarchy – obviously relevant as Europe was trying to settle down once more in the very last year of the Napoleonic wars – and about the unity of Denmark, island Danes and peninsular Jutes combined: Hrothgar’s alleged granting of “citizenship” to the Jutish plebs and senatorial status to their nobles does not come from the poem but from King Frederik VI’s contemporaneous attempts to win the loyalty of Schleswig-Holstein and especially of its troublesome Ritterschaft.47
And Thorkelin’s stance is clear from the title page, where the poem is announced as not only “DE DANORUM,” but also “POËMA DANICUM DIALECTO ANGLOSAXONICA.” The poem is conceived of as an assertion of pan-Danish identity, and it is clear from other parts of his biography that these ambitions were grandiose; that the great lexicographer of Scots, John Jamieson, aligned Scots with Norse has been connected with his encounter with Thorkelin during the Scottish part of the latter’s research expedition,48 with resonances for imaginative twenty-first-century attempts to identify a distinctive and historically situated “Nordic” Scottish identity. A Latin translation was itself a bold bid for a pan-European hearing – perhaps unwisely.
But Thorkelin’s edition does not express its ideology solely through the “paratexts” of his title page and preface and the Latin parallel text. His choice of roman font as opposed to “antiquarian” Pica Saxon may of course be constrained by what was available to his printer, Rangel of Copenhagen, but choosing to impose the half-line unit as the basic measure is a clear statement of a particular view of the structure of Old English verse. In doing so he was undertaking something that later editors, however critical they were of his efforts, were also to imitate with enthusiasm – thereby, of course, changing again the pragmatics of the text, guiding the ways in which that text was to be received.
Implications
Elsewhere, I have argued that
It is a truism of many disciplines that, when a cultural artefact comes down to us from the past – a poem, piece of music, painting, sculpture, tapestry – its “authenticity” as a witness for its own time may be remarked upon but it is also, of course, situated within twenty-first-century culture. A piece of “early” music is, for instance, just as much part of our contemporary cultural capital as a composition from our own time … And the ways in which (say) a medieval poem is presented in subsequent centuries relate dynamically to the changing ways in which the past is integrated within broader cultural/national narratives and imperatives. In sum, the present is always in dynamic dialogue with the past.49
These points are, I would suggest, relevant to the various case studies discussed here. What this chapter has attempted to demonstrate is that textual detail can be linked rather precisely to contextual setting; understanding the linguistic form of a text – in the broadest sense, bringing paleography/bibliography and linguistic research back into close articulation – relates closely to the sociocultural contexts in which that text exists.
