Old English Enigmatic Poems and Their Reception in Early Scholarship and Supernatural Fiction

Abstract

The scholarly reception history of the Old English riddles and adjacent “enigmatic poems” of the Exeter Book reveals a long process of creating intelligibility and order out of a complicated and obscure manuscript context. Understanding this history of reception allows us to see the influence of Old English poetry on modern creative medievalism, including the unexpected influence of medieval “enigmatic” poetry on the modern genre of supernatural fiction. Specifically, it is argued that the scholarly reception of folios 122v–123v of the Exeter Anthology was instrumental in inspiring one of the acknowledged classic ghost stories of the twentieth century, M.R. James’s “Oh, Whistle, and I’ll Come to You, My Lad”.

Our eye drawn to so many of his other landmark publications, it would be easy to overlook the great significance of Jack Niles’s contributions to the reception history of Old English literature.1 This research has left a truly invaluable legacy, however, for to sharpen perceptions of a field’s past is to release present possibilities. A long view is always useful, and it is all too easy to discount what has come before. In fact, as David Matthews has emphasized in his work on the history of medievalism, older medieval studies are always in peril of falling back into abject antiquarian oblivion, so that yesterday’s serious scholarship is perceived today to be mere fantasy, as fanciful as fiction—except in a genre very few want to read.2 When Jack encounters even the most dated, outmoded scholarship, however, it inevitably inspires in him not derision, but rather respectful consideration and reevaluation, and often, even, reinvention. Cast-off ideas summon something new.

Indeed, one paradox of medievalism—”the inspiration of the Middle Ages in all forms of art and thought”, in the words of Leslie J. Workman—is that the past must first be invented before it can influence.3 Such invention is always ongoing and accumulating its own history, as Bill Herbert’s modern Exeter-style riddle suggests: “I do not have a body/ yet I grow constantly”.4 Nearly any medieval text we might name has gathered much on its postmedieval path, beginning often with a narrow escape from the scrapheap. In the best of cases, the contents of a manuscript will be preserved, catalogued, edited, emended, titled, attributed, translated, annotated, and entombed beneath dozens of uneven layers of academic commentary. As a rule, only after many such scholarly interventions does inspiration of an avowedly “creative” kind arrive.

Yet many studies in medievalism neglect the significance of these contexts, even sometimes in cases where the painstaking work of medieval studies and the fantastical creativity of medievalism are obviously inseparable. The most obvious such case is the fantasy fiction of the Oxford professor J.R.R. Tolkien (1892–1973), but there are many others we might name. In comparison to Tolkien, the “antiquarian” ghost stories of the Cambridge provost M.R. James (1862–1936) have received far less attention from medievalists, though his tales have been very influential, especially in their use of artful reticence to arouse a response in the reader—that “pleasing terror” for which his tales are so celebrated.5 Tolkien’s debt to medieval riddling is well known, especially in the “Riddles in the Dark” chapter of The Hobbit, and the great popularity of that book ensures that many who read the Exeter Riddles today, whether in the original or in translation, do so with an understanding of the genre shaped in part by the sensibilities of Tolkien.6 Yet Tolkien’s creative riddling is itself best understood as in dialogue with medieval sources, not as fixed objects, but as informed by particular and always-shifting scholarly contexts.7 The same holds for M.R. James and his most famous and influential story, “Oh, Whistle, and I’ll Come to You, My Lad”, which here I will argue likely owes much overlooked inspiration to the “enigmatic poems” found in folios 122v–123v of the Exeter Book (Exeter, Cathedral Library, MS 3501), the sections containing what today we call Riddle 30b, Riddle 60, and The Husband’s Message.8 In James’s day though, this stretch of the manuscript tended to be read quite differently than at present, and to best assess his ghostly excavation of Old English enigmatic poetry, we must pay close attention to that past as well.

1. Early Scholarship: Obscurities Made Intelligible

It might not seem worth pointing out that what we name the Exeter Riddles, by definition and design, are enigmatic. From their earliest scholarly reception, though, a very elastic sense of that word has tended to blend with other sources of obscurity and unintelligibility quite distinct from the strategies of obfuscation and misdirection which are characteristic of their apparent genre. In fact, in the early days of their recovery, it was not even clear that a unified and coherent collection of Old English riddles was what scholars had on their hands. The deeper riddle was how to make basic sense out of the baffling last sections of the manuscript. In a book published in 1826, John Josias Conybeare pointed to the only—and therefore exceptional—Latin text found in the Exeter Book as representative of the challenge: “The obscurity attaching itself to much of this part of the MS. will be rendered most conspicuous by the following specimen of corrupt Latinity, which appears absolutely unintelligible”.9 Sixteen years later, in his 1842 editio princeps of the Exeter Book, Benjamin Thorpe offers a tortuous apology that echoes the key words of Conybeare’s statement: “Of the ‘Riddles’ I regret to say that, from the obscurity naturally to be looked for in such compositions, arising partly from inadequate knowledge of the tongue, and partly from the manifest inaccuracies of the text, my translations, or rather my attempts at translation, though the best I can offer, are frequently almost, and sometimes, I fear, quite, as unintelligible as the originals”.10 For the rest of the century and beyond, variations on this formula were to appear time and again, with many evasive sources of unintelligibility and obscurity expressed.11 This is not surprising. The Exeter Riddles have long offered a readymade metaphor for the challenge of their own study.

Rendering that multidimensional obscurity intelligible has been the work of more than two centuries, in order to arrive at our dominant contemporary image of the Exeter Riddles as a delimited collection of nearly a hundred enigmas complete with consensus solutions, identifiable links to multiple riddling traditions, generic and sectional differentiation from other poems in the manuscript, and even, possibly, a structural and thematic coherence mappable onto “encyclopedic principles of order”.12 Today, this orderly vision of the Exeter Riddles is often encountered as a tidy package of numbered texts isolated from most of the other poems of the Exeter Anthology. Niles has recently offered much illumination by simply reminding readers to take seriously the connections these texts have with the rest of the manuscript in a network of “horizontal” relationships.13 Those links are not limited to poems that also produce what Niles calls “bewilderment effects”—such as Wulf and Eadwacer and The Husband’s Message—though these are texts that especially blur the line between riddles proper and poems that are merely bewildering.14 Even sectional divisions in the Exeter Book remain in doubt, as evidenced by the various competing systems for numbering the individual Riddles.15 In fact, in the name of toppling artificially imposed barriers, some have recently endorsed the idea of removing numeration altogether and replacing the imposed title formula [Riddle + number] with something more individual for each poem. In my view, the practical wisdom of such a measure is questionable, but propositions such as this serve as an important reminder that what we make of the Exeter anthology of poetry is not a given—and not unchangeable over time.16

In fact, the earliest postmedieval scholars to study these texts do not seem to be primarily interested in them either as poems or as riddles. It is rather for their use of runes that they are included as specimen examples in the second volume of George Hickes’s treasury, or Thesaurus, of northern European medieval materials, published between 1703 and 1705.17 Hickes’s work, groundbreaking for early medieval studies, was accomplished under conditions of considerable personal distress and in collaboration with scholars such as the autodidactic marvel Humfrey Wanley (1672–1726), whose descriptive catalogue of medieval manuscripts (Catalogus Historico-Criticus) included in the Thesaurus was foundational to the particular field M.R. James was later to make his own.18 Hickes’s Thesaurus has left a tremendous legacy to Old English studies, for were it not for his transcriptions, many treasures would have been lost to fire or other destructive forces. The Old English Rune Poem, for example, is preserved only in this treasury, where there is much evident interest in runes in the work’s abundance of facsimiles and tables of variant characters drawn from a range of epigraphical contexts. Compiled in an “antiquarian” era predating the cordoning off of strict disciplinary bounds, Hickes’s Thesaurus freely gathers together its medieval materials, so that sections devoted to language and literature share space with studies of numismatics and manuscript illumination. To a degree not found in much later scholarship, considerable emphasis in the Thesaurus is placed on reproducing textual layout and letter forms with precision as they appear in their original inscriptional contexts, whether on parchment, coins, or the surface of objects that “speak for themselves” such as the Alfred Jewel or the Sutton Brooch, both of whose first-person inscriptions are carefully reproduced as facsimile illustrations within the Thesaurus.19

Likewise presented in facsimile are nearly all those Exeter Riddles that feature runic elements. These texts are not, however, to be found in the first volume of the Thesaurus, where Hickes provided what amounts to an anthology of Old English poetry in tandem with a grammar of the language.20 Some poetry from the Exeter Book (in particular, the Maxims21) is included there alongside other poems, but no riddles appear until the beginning of the second volume, not as supplements to the study of Old English language and literature, but rather as specimen illustrations accompanying an expanded version of Runolfur Jónsson’s 1651 Icelandic Grammar. The specimens selected—including portions of Riddles 19, 24, 36, 64, and 75–76, alongside the runic sections of The Ascension and Juliana (later to be identified as the “signatures” of Cynewulf)—are clearly chosen for their use of runes.22 The fact that these Old English riddles are represented in the incongruous context of an Icelandic grammar is not so surprising when we consider that scholars of this era generally assumed that runes were the “ur-script” of an ancestral Scandinavian language, one cloaked, from their perspective, in an “aura of quasi-pagan mystery”, as Niles puts it.23 These riddles, then, may have seemed to Hickes an especially good illustration because they used runes in such a bewildering way.

It is, in fact, difficult to separate Hickes’s interest in mysterious runic characters from his fascination with the voices he found speaking in the Exeter Book, as “not only the letters but also the voices are truly runic, that is mystical and occult”.24 The “runic” quality of these texts, then, is defined not just by their use of runes, but also by the enigmatic first-person voices of speakers who declare themselves either to be (Ic eom) or to have seen (Ic seah) something wonderful. All five of Hickes’s riddle specimens are headed by a large capitalized initial “Ic”, pronouns that are rendered even more prominent by being extracted from their original context and stacked together in facsimile. In an accompanying commentary, Hickes summarizes each of these puzzles, but does not attempt to solve them. For example, he explains that the text known today as Riddle 24 “also describes as a monster some thing or person, whose name is enigmatically set out in the runes”.25 Yet although he describes each of his runic examples as operating ænigmaticè, “enigmatically”, Hickes nevertheless does not appear to regard these texts as individual riddles so much as the components of a larger piece of religious poetry organized around a single enigmatic speaker: “But he who describes all these things, especially the sights of so many of them so mystically, is the dramatis persona, who also says many things about herself”.26 Hickes goes on to stitch together a sequence of additional Ic-statements drawn from other Exeter Riddles, including the opening few lines of Riddles 6, 8, 11, 15, 17, and 27. Presented with the puzzle of how such incongruities might be merged to apply to a single speaker, Hickes declares the dramatis persona to be evidently Ecclesia, the best candidate to embody such an amazing convocation of voices.27

Scholars today make distinctions among these speakers, so that Hickes’s Ecclesia doing battle for Christ is taken to be the Sun, while Hickes’s Ecclesia suffering persecution becomes a badger (or perhaps a fox or porcupine) harried in her den. But guessing the identity of these “riddle creatures” only began more than a century after the publication of the Thesaurus. Before this guessing game could begin, scholars had to first demarcate this section of the Exeter Book as a stand-alone collection, comprising a number of individually distinct and solvable riddles with answers as likely to be mundane as mystical. Hickes’s presentation does not encourage such recognition or supply complete texts for evaluation. The first sporadic solving efforts would need to be undertaken by those few who could access the Exeter manuscript in person. This John Josias Conybeare (1779–1824) was able to do in his relatively short tenure as the Rawlinsonian Professor of Anglo-Saxon at Oxford University, during which time he engaged with the Exeter Book in unprecedented depth. After his untimely death in 1824, Conybeare’s brother, William Daniel (1787–1857), himself a pioneering scholar (in the also emerging field of geological sciences and the fossil record), took up the task of continuing and completing for publication these researches, which appeared as Illustrations of Anglo-Saxon Poetry in 1826. The Illustrations is a landmark publication for a number of reasons, including its groundbreaking technical insights into the nature of Old English meter.28 Yet, simply in their presentation of texts and translations, the brothers Conybeare made a profound contribution to the field, making many Old English poems accessible to contemporary readers for the first time.

The scholarly debate around Old English riddle solutions also effectively begins with the Illustrations. Texts and translations for Riddles 3, 32, 46, and 66 are offered, along with proposed solutions of “sun”, “waggon or cart”, “Adam, Eve, two of their sons and one daughter”, and “the omnipresent power of the Deity”.29 Two of these four answers still enjoy at least some favor today, but more notable here is the Conybeares’ early recognition of the miscellaneous character of the Exeter Riddles: the riddle of the homely cart standing side by side the spiritual enigma of divine power. This mixing of the everyday with the mysterious has subsequently come to be seen as a central feature of Old English riddles, a standout characteristic of the genre. Indeed, the Illustrations is a work particularly interested in the matter of genre classification, beginning with an “Arranged Catalogue” of extant poetry divided into classes, including a catch-all category of obscure texts containing, among others, Widsith (here referred to as “the Song of the Traveller”), the Rune Poem, the dialogues of Solomon and Saturn, and the “Ænigmatical Poems” of the Exeter Book, referred to in discussion as both “ænigmata” and “riddles of the olden time”.30 This may be the earliest instance of classifying these texts with the vernacular terms “riddles”, and they are declared to be “so extremely obscure that they might suffice to damp the perseverance of a Saxon Œdipus”.31 Yet here, the thesis of an Ecclesia speaker is not wholly abandoned (Conybeare notes, “Others of the ænigmata appear to relate to the Christian Church, according to the opinion of Hickes”), so that the resolution of the Exeter Riddles into a singular generic category (let alone a unified, numbered collection) is not quite, at this point, fully affected.

This step was not taken until the editio princeps of the Exeter Book prepared by Benjamin Thorpe in 1842. In a field still dominated by the enthusiasm of untrained amateurs, Thorpe was a philologically trained professional scholar, and his numerous expertly produced editions have proven a rich legacy. Many of the editorial conventions he established still endure, and his standard of care and accuracy are relatively high for this era.32 For instance, although his representation of the runes of the Exeter Riddles were not able to match the high fidelity of facsimile possible in Hickes’s Thesaurus, Thorpe nevertheless faithfully reproduced the basic runic characters as he found them, rather than transcribing them into Latin equivalents, as we find in so many successive studies and editions. His translations—the first in modern English for the lion’s share of the Exeter Book, including most of the Riddles—were predictably flawed but still invaluable for offering access to many new readers. Thorpe’s sense of decorum, however, leads him to leave several of the more sexually charged riddles untranslated, including the notorious onion, key, dough, and churn riddles, along with the puzzle of the cock and hen (which combines elements of cryptography with barnyard copulation). He does translate a few riddles with similar, if slightly less obvious implications, though for one of these, a rare interpretive endnote seems calculated to stamp out any such unseemly possibilities, as noted for Riddle 21: “By this, no doubt, a plough is intended”.33

This same endnote may also offer an indication of the difficulties Thorpe faced in making ordered sense out of the contents of the Exeter Book. The endnote refers to the plough riddle as “Riddle XXI”, but the reference is quite plainly to the passage he has numbered “Riddle XXII” in the body of the text.34 The most obvious explanation for this mismatch is that Thorpe changed his mind at some point in his process, shifting his numbering to accommodate, as a riddle, the text known today as Wulf and Eadwacer, which he presents here as “Riddle I”. This first “riddle” is also left untranslated, though in this instance, not for reasons of decorum, but simply because it is too difficult: “Of this I can make no sense”, he notes in the commentary (one of Thorpe’s virtues as an editor was to admit when he was stumped).35 The most sense Thorpe could make of this text’s obscurity was to assign it the genre category of riddle, as that designation alone provided intelligibility of a kind. Moreover, simply adding it to the count also had the effect of rounding off the first group of riddles found in the Exeter Book to an even sixty. Thorpe indeed is more faithful to manuscript sequence than most subsequent editors, so that he begins the numbering anew with the next group (which also receives the numbering I-III) and yet again with a third group beginning I-III and so on. While this makes for a potential confusion of reference, it does have the effect of packaging the riddles as three neat sets of 60 (I-LX), 3 (I-III), and 30 (I-XXX). After all, medieval Latin enigmata are typically arranged in series with a round count. To create comparable order in the Exeter Riddles, Thorpe is obliged to identify as stand-alone riddles certain sections of text ambiguously presented in the Exeter Book (including what is now commonly interpreted as the opening of The Husband’s Message, as I will discuss further below). One might even say Thorpe’s editorial dilemma here is reminiscent of the plough’s neighbor, Riddle 22, with its sixty star-like riders—a puzzle of counting in its own right. However one strains, though, it is difficult to split or lump the collection in such a way as to produce the ultimate magic number of one hundred. Yet Thorpe’s split arrangement of the Riddles at least offers the honesty of reflecting their discontinuous distribution in the manuscript.

Thorpe’s edition put scholars in a much better position to respond to the Exeter Riddles with plausible solutions. The plough answer of Riddle 21 was Thorpe’s only guess and indeed, at that point, only a handful of other answers had reached print.36 But now the gates opened to a fertile new field of speculation, and many scholars have since lent a hand. Yet it was a single solver, Franz Dietrich, who was responsible for an impressive share of the first harvest. In two articles published in 1859 and 1865, Dietrich drew on Latin parallels, medieval material culture, and his own exceptional wit to solve nearly all of the Exeter Riddles. Well over half of the answers he offered have stood the test of time.37 To anyone familiar with the collection this informal list will speak for itself: storm, bell, swan, nightingale, cuckoo, leather, horn, badger, anchor, sword, bow, jay, onion, mead, moon and sun, bagpipes, iceberg, rake, mail-coat, bellows, bull-calf, creation, cock and hen, soul and body, key, bookworm, bookcase, battling ram, web and loom, swallows, well, chalice, reed, shirt, borer, beaker, another onion, another riddle of creation, ice, spear, oyster, ore, water, fish and river, one-eyed seller of garlic, inkhorn, another key, and another inkhorn. Though not every answer Dietrich offered has found favor, by the end of the nineteenth century his legacy was clear. In 1912 A.J. Wyatt wondered at the way Dietrich “by an effort of sympathetic imagination” had been able “to see and think with the eyes and mind of an eighth-century Englishman”.38

But Dietrich also made a contribution to perhaps the most distracting turn in the history of Exeter Riddles scholarship: the tortured chain of reasoning that led so many to assign them to Cynewulf as their wandering author. In 1840, two years prior to Thorpe’s edition of the Exeter Book, John Mitchell Kemble (1807–1857) had established that the runes of The Ascension and Juliana (both poems that appear in the Exeter Book) were to be deciphered to reveal the “signature” of Cynewulf.39 These passages, as I mention above, had long been closely associated with the runic Exeter Riddles, appearing side-by-side with them in composite facsimile in Hickes’s Thesaurus. In 1857, however, the German scholar Heinrich Leo took the association several unsupportable steps further by “solving” Thorpe’s first “Riddle I” (Wulf and Eadwacer) as a highly contrived charade-like puzzle concealing another Cynewulfian signature.40 In the opening enigma of the collection, then, Cynewulf was apparently claiming authorship over all the Exeter Riddles. The idea caught on and was swiftly linked to Dietrich’s 1859 proposal to solve the last riddle of the collection as “wandering singer” (fahrende sänger).41

Thus, Cynewulf the Wandering Singer of Riddles was born in the scholarly imagination and was soon to be encountered roaming through popular accounts such as Stopford A. Brooke’s The History of Early English Literature (1892), where we are invited to:

… imagine a wandering singer coming through the untilled woodland to one of the villages, to sing his songs, and to pass on to another. […] Then, our wandering singer (whom I will now call Cynewulf, because all the illustrations of village life which I shall quote are from his riddles), listening, heard the rushing of the water past the wattled weirs built out from its sides for the fishing, and saw the bridge of wood that crossed it, and perhaps mills by its side that ground the corn of the settlement, and thinking of the millstone made it the subject of his fifth riddle.

42

Brooke’s Cynewulf is a forerunner of the Romantics and the Exeter Riddles are the exuberant nature poetry he wrote in his youth (composed at the age of about twenty-five, before some unknown downturn in fortune darkened his subject matter and genre preferences43). In reference to what he takes as the badger of Riddle 15, for instance, Brooke notes:

It is in these short poems—in this sympathetic treatment of the beasts of the wood, as afterwards of the birds; in this transference to them of human passions and of the interest awakened by their suffering and pleasure—that the English poetry of animals begins. […] His sympathy is even more than that of Shakspere in his outside description of the horse or the hare. The note is rather the note of Burns and Coleridge […].

44

Brooke goes further than most to extract from the Exeter Riddles not only a name but a local habitation (“he was well acquainted with a storm-lashed coast”45) and indeed a full personality and turbulent biography. Not all accounts were quite so fanciful, but the Cynewulfian theory of riddle authorship enjoyed widespread acceptance in the scholarly world, retaining adherents decades after its premises were dismantled.46

But by the turn of the twentieth century the focus had decisively shifted away from attempts to detect secret Cynewulfian messages, settling instead on many of the concerns that continue to absorb scholars today. Certain aspects of the Riddles have proved to be of perennial interest. For instance, early scholars’ fascination with “runic voices” is matched by much recent discussion of the Exeter Riddles’ affordance of first-person subjectivity to speaking objects, work that is informed by a wave of theoretical interest in all things nonhuman.47 Such approaches frequently link the voice-bearing Riddles to other enigmatic first-person inscriptions on early medieval artifacts in a way rather in keeping with the textured, multidisciplinary approach we find in the Thesaurus.48 Of course the contemporary “object-oriented” turn in the field has been applied to a wide range of early medieval materials, but the Exeter Riddles have provided both a key example and a guiding metaphor. Indeed, James Paz lays out the scope of his larger study of Nonhuman Voices in Anglo-Saxon Literature and Material Culture (2017) in just such riddling terms:

The argument that I draw from these areas of focus is that,

although things are endowed with voices in Anglo-Saxon literature

and material culture, they also have an agency apart from

humans. This agency is linked to: one, their enigmatic resistance,

their refusal to submit to human ways of knowing and categorising

the world; and, two, their ability to gather, to draw together,

other kinds of things, to create assemblages in which human and

nonhuman forces combine. Anglo-Saxon things speak yet they can

be stubbornly silent. They can communicate with humans but, like

riddles, they also elude, defy, withdraw, from us.

49

A century and a decade earlier, Mary Bentinck Smith (1864–1921)—at the time, Director of Studies and Lecturer in Modern Languages at Girton College—would also highlight riddling encounters with the nonhuman, which she links to paganism and a more sinister sense of English landscape:

“in [the Old English riddlers’] hands inanimate objects become endowed with life and personality; the powers of nature become objects of worship such as they were in olden times; they describe the scenery of their own country, the fen, the river, and the sea, the horror of the untrodden forest […]”

50

Smith’s discussion here appeared in her contribution to the first volume of the multi-authored Cambridge History of English Literature (1907), where discussions of the Exeter Riddles are to be found in four separately authored chapters. Smith also remarks on the “peculiarly English tone and character of the riddles”, and her colleague at Cambridge, M.R. James, would seem to agree, writing in his directly subsequent chapter: “That this form of wit-sharpening made a great appeal to the mind of our ancestors is amply evident from many passages in the Old English literature”.51 Taken together, the contributions to this Cambridge History provide a useful snapshot of the consensus of academic opinion at the beginning of the twentieth century, just prior to the publication of the first separate scholarly editions of the Exeter Riddles by Frederick Tupper, Jr. in 1910 and A.J. Wyatt in 1912. Indeed, with Tupper’s edition in particular—still of considerable scholarly value today—one might well say that the Exeter Riddles as we know them had arrived.

But arrivals always come from somewhere. Early scholarship, I would like to suggest, not only rendered these texts intelligible, but also enhanced their dark resonance. An expert would be alive to such associations, even when obsolete, contradictory, or incongruous. For instance, turn-of-the-century readers versed in the scholarly dark matter surrounding the Exeter Riddles would be aware that Hickes’s occult runic voice—so suitable for illustrating the shadowy character of an Icelandic Grammar—is also that of a quintessentially English genre, rooted in national sensibilities and a familiar landscape. Both qualities coincide, along with other shadowy companions. As an heir to the “antiquarian” past, the scholar and storyteller M.R. James would be intimately familiar with the way the past grows as it gathers itself. We can turn now to his enigmatic ghosts.

2. M.R. James and the Voice of the Whistle

Ghost Stories of an Antiquary appeared in 1904, its publication motivated in part to benefit a friend, James McBryde (1874–1904), who was to provide illustrations. When the young artist died of appendicitis before completing this work, James’s small volume of stories became a memorial, dedicated to McBryde and “to all those who at various times have listened to them”.52 As the book’s title hints, nonacademic imaginative work of this kind was something of a departure for a scholar like James, who had by that time achieved an outstanding reputation in the scholarly world. His publisher Edward Arnold sought to trade on that reputation in advertisements: “Those who know the extensive and miscellaneous character of Dr. James’s researches in various fields of learning will not be surprised to find him appearing as the author of a volume of ‘Ghost Stories’”.53 The “antiquarian” accomplishments of James are difficult to summarize succinctly—the subject matter of his major publications ranged from biblical apocrypha, to hagiography, to art history, to church architecture.54 His most profound legacy, however, was the contribution of the many descriptive catalogues of medieval manuscripts he produced over his lifetime. In many ways, in fact, James was a professional heir to Humfrey Wanley and the Catalogus Historico-Criticus of Hickes’s Thesaurus.55 In his early thirties, James had been declared by Lord Acton (1832–1902), “in knowledge of MSS […] already third or fourth in Europe”.56 By the time James began publishing ghost stories, his academic reputation was more than well established. In 1903, he had been elected Fellow of the newly-formed British Academy (numbering among only the second cohort of inductees to be so honored) and received a congratulatory letter of “genuine satisfaction and delight” from Israel Gollancz, his one-time colleague at Cambridge (who had, incidentally, published the first volume of an edition and translation of the Exeter Book in 1895).57 And so, although James was not a specialist in the field of Old English poetry, his professional work was certainly engaged with the study of early medieval books, Old English manuscripts, and related artifacts.58 He would not need to look far afield for sources of antiquarian inspiration.

As is the case for many of his tales, “Oh, Whistle, and I’ll Come to You, My Lad” was first written for a Christmastime audience of intimate of friends and colleagues in James’s residential rooms at King’s College. These gatherings were festive, all-male affairs, where eerie ghost-story telling often gave way to raucous horseplay: “after which those played animal grab who did not mind having their clothes torn to pieces & their hands nailscored” (in one witness account of such a game of “animal grab”, we read of a participant thrashing about on the floor, with “Monty James’s long fingers grasping at his vitals”).59 We can imagine something of this atmosphere on the occasion of the probable first reading of “Oh, Whistle” in December of 1903, which was attended by guests including, among several others, McBryde and A.C. Benson (Benson was one of a number of regular attendees who would take to writing his own ghost stories influenced by the Jamesian style). In a diary entry describing this reading, “Oh, Whistle” is characterized by Benson as “one of [James’s] medieval ghost stories”, notable for the excellence of its “local colour”.60

The local color of this ghost story includes several medieval elements, including the darkly suggestive setting of a ruined Templar preceptory on the windswept Suffolk coast, where the tale’s protagonist, the vacationing Professor Parkins, unearths in a medieval ruin a haunted whistle through which he ventures to blow a note—a sound carrying “a quality of infinite distance”.61 This act whistles up the wind and brings to the professor’s bedroom a terrifying visitor, which shoves its horrible face into Parkins’s own in a terrifying semblance of a kiss. It was, after all, an unwise investigation from the beginning, for Parkins is no scholar of antiquities. His area of expertise is “ontography”, a term which James clearly invented in order to spoof what he saw as preposterous and arrogant fields of modern academic study.62 Curiously, and with no little irony, this same term has now been directly borrowed from James’s tale by the very object-oriented theorists I discuss above who have been so influential on the recent study of the Exeter Riddles and those poems’ representation of nonhuman being, agency, and voice.63

At any rate, with its richly atmospheric quality of fright, this story of Parkins’s encounter with the alien spirit of a haunted whistle has often been regarded as James’s finest, where he was able to “refine the essence of the ghost story” into “something altogether stranger and more frightening” than what had come before.64 It is a story we might well point to as a turning point in James’s approach to the genre. The celebrated subtlety of the Jamesian horror—rising with an undefined sense of unease and punctuated by “the very highest calibre of jolt”—is not quite as present in his earlier efforts, such as “The Ash-Tree” (first read in 1899, with witch-trials and spiders the size of kittens) and “Count Magnus” (first read in 1902, with its grisly gothic horrors and a sarcophagus sealed with “massive steel padlocks”). Before “Oh, Whistle”, arguably, James’s frights are rather more raw, bloody, and what James might call “blatant”.65 James came to recognize restraint as essential to the genre and his foundational example has become a model for those participating in the “antiquarian” ghostly tradition ever since: “Reticence may be an elderly doctrine to preach”, he wrote in 1930, “yet from the artistic point of view I am sure it is a sound one. Reticence conduces to effect, blatancy ruins it, and there is much blatancy in a lot of recent stories”.66 The story of the haunted whistle is certainly reticent in its implications, so much so that afficionados online continue to speculate on the backstory of a whistle unearthed in a ruin. How did it get there? Most would concede that James was right to leave us in the dark on this point. Yet despite all the ink shed—some of it mine—annotating and explicating this extensively anthologized tale, one particularly promising medieval source of inspiration has gone unnoticed.67 Readers familiar with the Exeter Book—and in particular, the Old English enigmatic poem The Husband’s Message—may now be ready to guess where this discussion is headed.

The basic parallel of a speaking object declaring its beckoning message by the sea is only a beginning of what potentially links these texts, and the discussion below will point out many further particulars. To begin, however, note that The Husband’s Message centers around an enigmatically inscribed object arriving on the coast from a faraway land (eom nu her cumen), reminding its recipient of an eald gebeot, “old pledge”, and summoning her to a lover who eagerly awaits her coming on wenum, “in expectation”. The title of James’s story alone might remind us of this situation, though its exact wording is borrowed from another source: the refrain of a 1793 song by Robert Burns:

• O Whistle, and I’ll come to you, my lad;
• O whistle, and I’ll come to you, my lad:
• Tho’ father, and mither and a’ should gae mad,
• O whistle, and I’ll come to you, my lad.
68

The larger contexts into which this line is imported has led many to suspect a subtext of gender anxiety or same-sex desire in James’s tale, a story which, after all climaxes, with a bedsheeted specter pressing its face suggestively into that of Parkins.69 Such scenarios of terrifying intimacy were to become a hallmark of the Jamesian approach to horror writing, especially in the tales he wrote after “Oh, Whistle”. Take, for example, “A School Story”, composed in 1906 and published in his follow-up 1911 More Ghost Stories of an Antiquary, which concludes with a haunted victim hauled off to the bottom of a well, embraced there eternally by a persecuting specter who had announced his imminent arrival with cryptic words: “Si tu non veneris ad me, ego veniam ad te”, “If you don’t come to me, I’ll come to you”.70 In the case of Parkins, the professor’s scare is—at least on the surface—a punishment for his stuffy incuriosity and arrogance, though it can also be interpreted as a phobic manifestation of other sources of anxiety.

The more immediate cause of the haunting, however, is simply a failure to construe the meaning of the inscriptions found on the whistle, which are worth reproducing as they appeared in the first edition of Ghost Stories of an Antiquary:

M.R. James, Ghost Stories of an Antiquary (London: Edward Arnold, 1904), p. 199.

Photo by the author, from a copy of the first edition, Eton College Archives, Lq.4.06.

The meaning of this puzzle has long been disputed by Jamesians, partly because the uniquely bracketed swastika-like fylfot crosses flanking the second inscription were substituted out for the plain ones in James’s Collected Ghost Stories (1931), as well as in all subsequent editions until very recently. To sum up a rather complicated situation, it is now clear that the bracketed arms of the second inscription were intended (confirmed by witness of James’s holograph manuscript of the story) to provide a key to the riddle of the first inscription, so that bis is linked with two of the three other enigmatic syllables. Thus, the first inscription can be construed, “Fur, flabis, flebis”, “O Thief, you will blow, you will weep”, in what amounts to a compact (and sinister) synopsis of the story that is to unfold.71

Like “Oh, Whistle”, “the Old English poem The Husband’s Message centers around a riddlic inscription with ambiguous characters to be rearranged and decoded. As Niles has noted, however, most modern editions of this Old English poem—especially those published subsequent to the early work of scholars like Hickes and Thorpe—have tended to efface and so suppress the ambiguity of the characters as they appear on the parchment of the Exeter Book by rendering them into plain Latin characters of S, R, EA, W, and D (or M). In the case of The Husband’s Message, this de-”runification” has contributed additional confusion to the interpretive history of an already difficult text.72 A parallel situation arose with the fylfots of James’s whistle inscription, which lost their brackets probably for similar reasons of printing practicality. The fact that both texts have proven vulnerable to such effacement is an unintentional coincidence, but not an insignificant one. As a scholar exceptionally immersed in the study of manuscripts and their great variability of form, James was keenly attuned to the unruly qualities of medieval textuality. The creation of his own whistle inscription with rune-like fylfots unsurprisingly reflects a sensitivity to the potential importance of spatial layout and ambiguous “runified” characters, features that prove resistant to modern reproduction.

The rest of the whistle inscription is transparent enough for even Parkins to translate, yet still offers a riddle-like challenge:

“I ought to be able to make it out”, he thought; “but I suppose I am a little rusty in my Latin. When I come to think of it, I don’t believe I even know the word for a whistle. The long one does seem simple enough. It ought to mean, ‘Who is this who is coming?’ Well, the best way to find out is evidently to whistle for him”.

73

A curious detail here is Parkins’s assumption that a medieval word for whistle might be relevant to construing the inscription. This suggests an awareness (whether we want to attribute that awareness primarily to Parkins or to James) of the penchant for medieval inscribed objects to name themselves, as does, for example, the Brussels Cross (“Rod is min nama”, “Cross is my name”) and a comb-case discovered in 1867 (“kamb: koþan: kiari: þorfastr”, “Thorfast made a good comb”).74 Often such inscriptions take on the voice of the object itself, as in the case of the ninth-century Alfred Jewel (“ÆLFRED MEC HEHT GEWYRCAN”, “Alfred ordered me made”75), or the Sutton Brooch (eleventh century) which, like James’s whistle, offers a warning to thieves:

AEDǷEN ME AGE HYO DRIHTEN

DRIHTEN HINE AǷERIE ÐE ME HIRE ÆTFERIE

BUTON HYO ME SELLE HIRE AGENES ǷILLES

(Aedwen owns me, may the Lord own her. May the Lord curse him who takes me from her, unless she gives me of her own free will”.)

76

It is this similarity of speaking objects that has led many scholars to link such inscriptions to the Old English riddling genre, which makes such effective use of prosopopoeia in first-person texts challenging solvers to “saga hwæt ic hatte”, or “say what I am called”.77 Of course, the inscription on James’s whistle does not speak in the first person, but only enigmatically inquires, “quis est iste qui uenit”, “Who is this who is coming?” Yet, the disembodied voice of the object does manage to speak for itself in the title of the tale, detached and floating ominously on the epigraphical edges: “Oh, Whistle, and I’ll Come to You …”78

This untethered voice, with its enigmatic relation to the runified whistle inscription, is a striking echo of what we find in The Husband’s Message. In the Old English poem, the voice bidding the beloved to come is evidently that of a wooden item (treocyn, line 2) speaking on behalf of its absent lord, but also presenting a more circumscribed message in the form of riddlic runes. Whatever we take the literal object to be (a runestick, the mast of a ship, or some other tree-made thing79), its postmedieval critical reception has revolved around the curiously disembodied and paradoxical quality of its speaking position (made even more obscure by burn damage to the Exeter manuscript), so that some readers have resorted to positing a split between the object and the separate voice of a human emissary delivering and providing a lengthy gloss for the inscribed object on-site.80 Most readers, though, have more simply attributed the entire monologue to the voice of the runic object, yet this only heightens the oddly disjointed relationship between the message of the runes and the disembodied voice of the poem’s treocyn speaker. As Ralph W.V. Elliott explains it:

… the five runes [are] the actual message supposed to have been carved into the wood and sent to the wife. They may represent a secret cypher previously agreed upon by husband and wife; in any case, it is clear that we cannot be expected to regard the whole seventy lines of the poem as having been inscribed on a runakefli. If this assumption is correct the poem may properly be deemed an explanation of the terse runic message in greatly expanded form. This expansion allows the inclusion of the wood’s own history as well as the more detailed exposition of the actual situation of husband and wife and the message sent by the former.

81

If some explanation along these lines is accepted, the poem is an unusually complex example of the “phenomenon of ‘voices within voices,’” notable as a curious feature of the Exeter Anthology.82 Of course, it is possible that such voices would not have been quite so bewildering to a medieval audience.83 Yet as “eavesdroppers” into this cryptically intimate communication—via a split voice at once carved into solid wood and yet mysteriously disembodied—many modern readers have found The Husband’s Message to be exceptionally obscure, even by the standards of this manuscript.84 And this quality may be what inspired James to create his own parallel object, complete with its own alien and disembodied voice that beckons menacingly from beyond.

But why has the Old English object become, for James, a whistle?85 A ready answer is revealed when we look not only at the Exeter poems, but also at their scholarly reception. As I have noted, “Oh, Whistle” was probably written around December of 1903. The timing here may be significant, as the immediately preceding years saw the publication of an important and controversial article by F.A. Blackburn in 1900 (making a splash large enough to receive an endorsement from and a full paragraph of summary in the 1907 Cambridge History).86 Blackburn’s article engaged anew with those questions of sectional division in the Exeter Book that had challenged scholars ever since Thorpe’s edition. It might well have caught the attention of James, a scholar whose daily work involved cataloguing the individual items contained within medieval manuscripts.87 Blackburn argued that the text of The Husband’s Message was, in fact, best interpreted as part of the text that immediately preceded it: Riddle 60 (or “Riddle 61”, according to the numbering of the Thorpe edition used by Backburn). As Blackburn noted, Riddle 60 shares much in common with The Husband’s Message, describing as it does, in the first person, the experiences of an enigmatic object that, having been shaped by human skill, allows for the conveyance of a secret message.88 Blackburn thus contends that “when we read the whole as a single poem, we find a consecutiveness and unity so clear” that it would be quite natural to read them together as a single continuous composition. Indications otherwise in the Exeter Book were explained as scribal error. Together, the two sections should be taken to form a single text, to which Blackburn proposes we apply the title of “A Love-letter”.89

Blackburn’s theory has not been wholly successful in persuading later editors to re-title and redraw the boundaries of these poems, but neither has it been altogether discarded.90 Certainly, the texts share much in common. After all, like the speaker of The Husband’s Message, the voice of Riddle 60 has been often taken to be a runestick/staff, a solution favored, for instance, by Moritz Trautman in 1894, and by Henry Morley in 1888 (in accounting for its seaside origins, Morley described the object as specifically a “letter-beam cut from the stump of an old jetty”).91 More widely adopted in 1903, though, was Dietrich’s proposal of rohrflöte (hvistle), or “reed-pipe (whistle)”, an identification largely based on its perceived similarity to Symphosius’s “Arundo”, or “reed” enigma.92 In fact, Frederick Morgan Padelford, in his book on Old English Musical Terms (1899), cited this very text to illustrate ancient English whistle nomenclature. If we were to imagine Parkins actually visiting a library to look up old words for whistles, Padelford’s volume would be the most obvious place to start:

Pipe and hwistle were also the names of instruments of the flute order, for tibicen is glossed as pipere oððe hwistlere, and auledus as reodpipere. The reed-pipe is the subject of the sixty-first riddle.

93

Like Parkins’s whistle, the speaking object of this riddle dwells alone, secured in a covert spot by the sea. The full Riddle 60 reads:

Ic wæs be sonde,   sæwealle neah,

æt merefaroþe   minum gewunade

frumstaþole fæst;   fea ænig wæs

monna cynnes   þæt minne þær

on anæde   eard beheolde,

ac mec uhtna   gehwam yð sio brune

lagufæðme beleolc.   lyt ic wende

þæt ic ær oþþe sið   æfre sceolde

ofer meodu[drincende]   muðleas sprecan,

wordum wrixlan.   þæt is wundres dæl,

on sefan searolic   þa þe swylc ne conn,

hu mec seaxes ord   7 seo swiþre hond,

eorles ingeþonc   7 ord somod,

þingum geþydan,   þæt ic wiþ þe sceolde

for unc anum twam   ærendspræce

abeodan bealdlice,   swa hit beorna ma

uncre wordcwidas   widdor ne mænden.

[My home was on the beach near the sea-shore;/Beside the ocean’s brim I dwelt, fast fixed/In my first abode. Few of mankind there were/That there beheld my home in the solitude,/But every morn the brown wave encircled me /With its watery embrace./Little weened I then that I should ever, earlier or later,/Though mouthless, speak among the mead-drinkers/And utter words. A great marvel it is,/Strange in the mind that knoweth it not,/How the point of the knife and the right hand,/The thought of a man, and his blade therewith,/Shaped me with skill, that boldly I might/So deliver a message to thee/In the presence of us two alone/that to other men our talk/May not make it more widely known.]

94

The central conceit here is the riddling motif of “mouthless” speech, and many have found in this text a clear reference to writing, whether in the form of runes inscribed on a slip of wood, or as penmanship accomplished with a sharpened reed. The paradoxes of written and spoken language are a well-studied theme of Exeter riddling, where the whispering quality of all writing often overlaps with playful games of runic concealment.95 Riddle 60 seems in tune with that theme, yet the clue of speaking among mead-drinkers is a point in favor of a musical instrument. An elegant way around this impasse, as Niles has established, comes when we “answer the riddles in their own tongue”, so that the single Old English word hreod can encapsulate the protean identity of the reed—a cylindrical creature that can “speak” both as pen and as whistle.96

Early solvers, however, felt compelled to make the case for “reed whistle” by way of an alternative explanation to account for the sense of secret communication we find at the end of the riddle. In the lively imagination of Stopford Brooke (1892), the theme of silent speech becomes clandestine romance, an intense intimacy conveyed by the sound of the whistling. But Brooke’s real interest lies in the way the theme merges with the English landscape in the riddle:

It tells of a desert place near the shore, traversed by a channel up which the tide flowed, and where the reeds grew which were made into the Reed-Flute, which is the answer to the riddle. I translate the whole. The picture, at the end, of the lover talking in music to his sweetheart, music that none understood but she, is full of human feeling, but the point on which I dwell is the scenery. It is that of a settlement where only a few scattered huts stood amid the desolate marsh.

97

The secret notes of the reed-flute, of course, easily mingle with the beckoning message of The Husband’s Message, once these texts are merged by Blackburn. In James’s hands, at any rate, the scene turns to horror, and countless critics have emphasized the atmospheric brilliance of the tale, set in the seaside resort town of Burnstow, a lightly disguised version of Felixstowe on the Suffolk coast.98 James’s word-painting in the tale is indeed lovely:

Bleak and solemn was the view on which he took a last look before starting homeward. A faint yellow light in the west showed the links, on which a few figures moving towards the club-house were still visible, the squat martello tower, the lights of Aldsey village, the pale ribbon of sands intersected at intervals by black wooden groynes, the dim and murmuring sea.

Such “local colour” noted by Benson at the tale’s first reading has subsequently been analyzed time and again in terms of its “agoraphobic sea horizons;” “the cumulative forces of the eerie that animate the East Anglian landscape;” the way it evokes “the windswept mystery of the barren unknown”.99 Such characterizations are in no way to be dismissed; James’s fiction is certainly rooted in an English landscape he personally experienced and found deeply evocative, but it is also informed by his engagements with medieval studies, and here the sense of the enigmatic looms large. Even the “shape of a rather indistinct personage”, the “bobbing black object” conjured on the shoreline by the whistle, behaves quite like an unresolved riddle creature, declaring its own incongruous form for the solver’s contemplation: “a little flicker of something light-coloured moving to and fro with great swiftness and irregularity. Rapidly growing larger, it, too, declared itself as a figure in pale, fluttering draperies, ill-defined”.100

The sense of enigmatic intimacy we find in both Riddle 60 and The Husband’s Message finds its analogues, too, in the celebrated culmination of James’s tale, as the whistle-summoned horror arrives in the loosely wrapped, sinuous form of a bedsheeted figure gliding across the bedroom to press in upon Parkins in his state of “horrid perplexity”. The climax brings a “face of crumpled linen” that is “thrust close into his own”.101 It might be worth comparing this specter with yet another Exeter riddle, the one that, in fact, immediately precedes Blackburn’s “A Love-letter”. As Blackburn argues, Riddle 30b is also likely a creature of wood, yet its duplication here (Riddle 30a is found earlier in the Exeter Book) seems to rule out the possibility that it, too, might be properly joined to what follows. Blackburn, in fact, argues that its repetition in this part of the manuscript can be explained as another error, this time a mistaken attempt to join text that does not belong to “A Love-letter”. If this is a scribal error, then it might also be a productive one in James’s medievalist ghost story, for the opening of Riddle 30b is very reminiscent of the flexuous creature conjured by the whistle, particularly if we look to Blackburn’s emended text:

ic eom licbysig,   lace mid winde

w[unden mid wuldre   we]dre gesomnad

[I am agile of body, I sport with the wind. I am clothed with beauty, a comrade of the storm].

102

Readers familiar with Riddle 30b will notice Blackburn has emended MS “ligbysig” (“flame-busy”, likely reflecting the capacity of wood to burn) with “licbysig”, “agile of body” (a reading in which Blackburn ultimately followed Thorpe, who translated the half-line as “I am a busybody”).103 James’s whistle ghost is also quite agile:

It would stop, raise arms, bow itself toward the sand, then run stooping across the beach to the water-edge and back again; and then, rising upright, once more continue its course forward at a speed that was startling and terrifying.

104

Such uncanny movements match the bewildering effects of Riddle 30b: “þon ic mec onhæbbe/hi onhnigað to me/modgum miltsum” (“when I rise up, before me bow/The proud with reverence”).105 Proud men also kiss the creature of Riddle 30b, a fate the arrogant Parkins nearly suffers before the amorphous, wind-sporting horror collapses into “a tumbled heap of bed-clothes”.106

It is true that James’s tale bears only a superficial resemblance to The Husband’s Message as it is usually encountered today. But when considered from the scholarly perspectives of 1903, it is easier to see how James may have found dark inspiration in Blackburn’s “A Love-letter”—with its composite features of an inscribed whistle concealed on the desolate shoreline, beckoning with an untethered voice for a lover to come calling, an enigmatic message associated with another amorphous and alien being of the wind. In James’s virtuoso remaking, though, these elements acquire a peculiarly haunting quality, and recognition of this inspiration affords insight into the affective power of the story. The claim—long controversial among some Jamesians—that “Oh, Whistle” taps into anxieties of sexual desire and homosocial intimacy is strengthened and enriched considerably by recognition of the tale’s connection to Blackburn’s “A Love-letter”. Additionally, appreciating James’s medievalist sources also allows us to better understand how such themes are linked in particular to scholarly experience in the tale, where Parkins’s antiquarian discovery is framed by questions of academic companionship, isolation, and alienation. James’s transformation of Exeter Book folios 122v–123v becomes a haunting and anxious meditation on the many ways men in such circles call out to one another—alluring and reassuring, desirous and threatening. There is a touch of the elegiac in this story of the self-exiled Parkins, who refuses the company of Cambridge colleagues on his stay by the sea.107 The alien voice of the whistle, then, takes on a new resonance when considered in these contexts: the enigmatic quality of the title itself and its strange relation to the “runified” object is a crucial aspect of the story that has surely been sensed by readers more often than explicitly identified as a source of the tale’s impact. And if “voices within voices” are a hallmark of Old English enigmatic poetry, they also seem to have helped shape the Jamesian style of “reticent” horror at a crucial moment both in James’s career and, therefore, in the development of the contemporary genre of supernatural fiction.

Of course, further connections are always a temptation. For instance, we might remark that Riddle 30b, Riddle 60, and The Husband’s Message are followed immediately in the Exeter Book by the damaged poem The Ruin. Is this why a ruin is where the hidden whistle is found? Perhaps, perhaps not, and the wisdom of pursuing these links much further is doubtful. Sometimes it is hard to know when to stop scraping at the turf.108 To be clear, I am not claiming that James stumbled across his reticent style of horror while rooting around in the Exeter Book, but engaging imaginatively with these enigmatic poems—as, crucially, they were received at this particular scholarly moment—may have allowed the self-styled antiquary to enhance his particularly effective and influential approach to horror. James brilliantly reinvented the spirit of the Old English enigmatic voice, and it was his own recognition of the riddle creature’s potential—its reticent resonance within the ghostly genre—that has proved to be such a source of “pleasing terror” for numberless readers. This alone is a rich legacy.

The role that Exeter enigmatic poetry played in the formation of the modern fantasy genre—especially by way of Tolkien—is very well-documented, but its part in the history of horror fiction is also worth acknowledging, considering the way so many later writers have emulated both “Oh, Whistle” and James’s approach to the genre in general. James’s “ontographical” response to the voice of the Exeter Book now appears even to have helped inspire theoretical interests that circle around to animate present-day medievalist scholarship on the Exeter Riddles themselves.109 This seems appropriate enough. A singular feature of these unsolved texts is the way they invite us to complete them creatively with our own voices, as we offer up imaginative guesses in a distantly familiar tongue. Of course it is rarely possible to banish all forms of fantasy from scholarship. Medieval studies and creative medievalism have always been joined together, even if the links are easily lost. This may be a disappointingly unparadoxical point on which to end, but I appeal to the inimitable example of a mentor who has offered so much illumination by pursuing deceptively simple ideas with uncommon seriousness and skill. I cannot match his light, but I have at least tried to avoid needless obscurity.