Another Gandalf who signed himself with a G.

Probably because of their success in Austin, food trucks have now arrived in Dallas. One of these is Gandolfo’s Deli, and this got me thinking about the name, Gandolfo, as a variation on Gandalf. I’ve seen it before — Castel Gandolfo, for one, is a town in Lazio, about a half-hour drive south of Rome, and the likely site of the fabled Alba Longa. I actually came within 100 km of this small town when I visited Italy in 2005. Most of our time was spent in Tuscany, but we made a couple of ventures into Umbria and Lazio as well — the latter, to the environs of Vacone, an even smaller località than Castel Gandolfo.

We know that Tolkien’s Gandalf has a Germanic name, specifically Old Norse. Tolkien borrowed the name from the Dvergatal section of the Völuspá (it also appears in the Ynglinga Saga, part of the Heimskringla of Snorri Sturluson [1]). I need not unduly repeat myself and others here (for one example, a long time ago, you might want to read this), but we can gloss the name as gandr “wand, magic staff” + álfr “elf, fairy”. So much is well known, but what of the name, Gandolfo? Is it related, or merely coincidentally similar? Should we expect Gandolfo to have a Romance etymology, rather than a Germanic one? If so, I can’t come up with anything plausible — anyone have any ideas? — so I am inclined to think it may have been borrowed into the Italic branch from the Goths, Franks, Lombards, Bretons, or another Germanic tribe of Late Antiquity. More on these etymological ruminations in a bit. First, back to Gandolfo.

What is its provenance? As with most things Italian, we start by looking directly to Latin. In this case, we would expect to find something like Gandolphus or Gandulphus. There are a number of historical figures with this name, and one in particular jumped out at me: Magister Gandulphus, a medieval canonist of twelfth-century Bologna (d. ca. 1185), and author of Sententiae. Gandulphus was a contemporary of the better known Peter Lombard (d. 1160), bishop, canonist, and author of Libri Quattuor Sententiarum. This was a much more influential work in medieval theology, one on which no less than Thomas Aquinas wrote commentaries a half-century later, but the Sententiae of Gandulphus were every bit as important in their day. In fact, “[t]he two works [appear] so similar in purpose, method, and content […] that some have been tempted to find the work of Peter Lombard in debt to the writings of the Bologna canonist” [2]. Some have even gone so far as to use the “p word” (plagiarism) — though to be fair, I’ve seen the same charges being made in the other direction as well.

Now I have no idea whether — and no particular reason to think that — Tolkien was familiar with this Gandulphus, not in the way we know he was of the semi-legendary Norse figure, but it is an interesting coincidence that the writings of Gandulphus were identified with a siglum. The signing of glosses and other writings with sigla is not unusual, but in this case, it does catch the eye. Gandulphus used various sigla in the extant writings — g., G., Ga., Gan., and Gand. are all recorded [4]. Of these, the siglum, G., is especially tantalizing, since it is also Gandalf’s sign. Readers of Tolkien will remember that Gandalf used the same siglum to mark his fireworks (in two different runic alphabets), to sign the letter he left for Frodo in Bree, and (apparently) to mark a stone on Weathertop.

So it appears that we have two personages with the name Gandalf (allowing for spelling), both signing their writings or otherwise identifying themselves with a G. I don’t mean to imply a direct borrowing here — with all that we know of the history of Tolkien’s Gandalf, it would seem unlikely that Tolkien had yet another source. Nor would he need a source to tell him that Gandalf should sign with his initial; that could proceed perfectly naturally from the character alone. But it is a surprising coincidence to find a real and a fictive Gandalf both doing this. The Norse Gandalf does not — not that we know very much about him.

So, to return to the etymology. The meaning of the first Norse element is, in fact, a bit unclear. It seems to have something to do with wizards, their magic, and/or the equipment by which they work it. It’s often calqued as “wand”, but this implies (incorrectly) that the Modern English reflex for ON gandr is actually “wand”. It isn’t. English wand is indeed borrowed from the Scandinavian branch (no pun intended!), but from ON vöndr “wand, switch, twig” — cognate to Gothic wandus < *bi-windan “to wind”. The second element definitively means “elf, fairy” and is a mainstay of Germanic anthroponymy (Alfred, Alvin, et al.). It is unattested in Gothic, but would have been something like *albs [3]. In theory, a third- or fourth-century Gothic or Lombardic name along the lines of *Wandalbus or some such could have been introduced into Latin as Gandolphus. What about the change in the vowel from front (a, æ, e) to back (o, u) — does this rule out a gloss of “elf”? No indeed. The Modern English oaf is one of several dialectal variations on elf. Drayton used the form aulf, Shakespeare ouphe. The idea was that an elf-child or “changeling” was sometimes left in place of a newborn baby, and that this child showed itself to be foolish, simple, or contrary-minded — that is, elf in the ancient sense became oaf in the modern sense.

It should be mentioned that there is an alternative theory. While most lexicographers agree that oaf derives from elf, it has been argued that oaf derives rather from auf “owl” (among its cognates, Old English úf, Old High German úvo, and Old Norse úfr “a bird of unknown kind”). Compare this to Italian gufo “owl”, and likewise compare French goffe “dull” and Modern English goof. This is one of a series of bird-names used metaphorically for silly, foolish, or mentally defective people — e.g., cuckoo, booby, dodo, gull (cp. gullible) — making it a pretty strong contender to explain the word oaf. But of course, this seems less likely to be the explanation for the second element in Gandolfo — unless Gandolfo was originally the name of a country clown or court jester.

So unless a reader can unearth an alternative explanation for the independent development of this name in the Italic family, I’m going to presume that it was most likely borrowed into Latin from one of the Continental Germanic tribes during the early centuries A.D., as were many other names, and that it has the same meaning as the Scandinavian Gandalf. That Tolkien’s Gandalf — a kind of angel, really — would bear some similarities to Gandulphus — not an angel, but a medieval Italian canonist [5] — is probably just an entertaining coincidence. But one, I think, worth spending a few words on. :)


[1] There are some eye-catching tidbits in the Ynglinga Saga as well (in addition to elements I discussed in my essay on Tolkien and the Heimskringla, which has been reviewed favorably). One passage, for example: “Olaf came to the kingdom after his father. […] He had Westfold; for King Alfgeir took all Vingulmark to himself, and placed his son Gandalf over it.” Olaf, though unrelated to elf, looks rather like it. Alfgeir is related to elf, and may be glossed “elf-spear”. Westfold, a part of the Vingulmark, of course, would cause any reader of Tolkien to sit up! There is also an Eastfold. But again, most likely no more than coincidence.

[2] John F. Sweeney, S.J. “Book Review of Le Mouvement Théologique du XIIe Siècle: Etudes, Recfarches et Documents, by J. de Ghellinck.” Theological Studies 11 (1950): 627–30, p. 628.

[3] William H. Bryson. Dictionary of Sigla and Abbreviations to and in Law Books Before 1607. Charlottesville: University Press of Virginia, 1975, pp. 72–3. See also Wilfried Hartmann and Kenneth Pennington. The History of Medieval Canon Law in the Classical Period, 1140–1234: From Gratian to the Decretals of Pope Gregory IX. Washington, D.C: Catholic University of America Press, 2008., pp. 73–4.

[4] This is one of the difficulties with Gothic. Its limited surviving lexis is skewed toward Christian words and away from the older pagan Germanic traditions. A sad loss in the native word-stock, one which would be repeated some five or six centuries later in England.

[5] There is a Saint Gandulphus as well, or more than one, distinct from Magister Gandulphus of Bologna. Of one St. Gandulphus, it has been written that “Many Persons derided his Miracles, and even his Wife scornfully told him, that he performed them just as she farted: Whereupon she violently broke Wind, and continued to do so, whenever she spoke a Word, on the same Day of every Week to her Death” (George Lavington. The Enthusiasm of Methodists and Papists Compared. Volume I. London: J. and P. Knapton, 1754, p. 202; italics original). Let that be a lesson never to insult the magic of a wand-elf!

Slavic echoes in Tolkien — A response

This may be old news, depending on how many of you read the Ukrainian literary journal, Літературознавчі студії (“Literary Studies”), published under the auspices of the Institute of Philology at the Taras Shevchenko National University of Kiev. Doesn’t everyone? Hahae, I know, neither do I. Not speaking Ukrainian is a bit of an obstacle, but hey, that’s why we invented machine translation. Thus armed, I have tackled an essay in the latest issue. Why? Well, because the author cites me, of course! Why else would I bother? Did I mention it’s in Ukrainian?! ;)

All kidding aside, the essay in question is “Слов’янські відлуння у творчості Дж.Р.Р. Толкіна” (“Slavic echoes in the works of J.R.R. Tolkien”), by Дмитро Кузьменко (henceforward, Dmitry Kuzmenko). For those who’d like to read it (with or without the help of robots), Kuzmenko has put a copy of the paper up on his website, here. It’s fairly short, a little under 2,000 words, but for those who’d like it even shorter, let me boil down the author’s argument.

Kuzmenko begins with the acknowledgement that Tolkien’s fictive world is one apparently “devoid of any influence of the Slavic culture and especially its literature”, but then asserts that “more detailed analysis shows otherwise” [1]. What does such analysis reveal? Kuzmenko spotlights several pieces of evidence:

• Tolkien’s invented language, Quenya, “has a few Russian words and suffixes”, which probably came about when Tolkien attempted to learn some Russian in 1918 [2].
• Kuzmenko asserts that “many [of Tolkien’s] toponyms […] were also words of Slavic origin”, such as Rhovanion; or if not necessary given Slavic names, they correspond to Slavic features, as he says the Anduin River springs from the Danube, “which flows mainly among the Slavic peoples and played an important role in their folklore”.
• The example of the wizard Radagast and his home at Rhosgobel, both usually taken to have Slavic sources, although “the exact etymology of the word ‘Radagast’ remains controversial”; it is here that Kuzmenko cites me, specifically this post written for Lingwë.
• The Variags of Khand, who (it can scarcely be argued) take their name from the Varangians, Norsemen who settled in the regions we now call Russia, Belarus, and the Ukraine.
• The original name for the character who would become Beorn in The Hobbit: Medwed, a Slavic name with the meaning “bear, honey-eater” (and which I also discuss at some length in the post on Radagast).
• Kuzmenko suggests that “several images [...] at first glance have a Germanic origin, but on closer examination reveal Slavic parallels”, and discusses three: Mirkwood, Wargs, and the dragon Smaug.

In the end, Kuzmenko’s conclusion, that Tolkien “consciously or unconsciously achieved a literary effect in terms of Slavic culture through the prism of the Germanic”, goes too far on too little evidence. Tolkien’s intentions are essentially set aside as irrelevant and the researcher’s own argument put forward in their place (“consciously or unconsciously”, he says; but doesn’t it matter which?). Better would be to say that Tolkien did indeed incorporate a few Slavic elements into the multifaceted and multi-sourced structure and background of Middle-earth, but that he did so quite sparingly and, in almost every case, only at its furthest margins. Kuzmenko also offers little in the way of original research. The examples he gives may nearly all be found in the works he cites, most significantly in Robert Orr’s 1994 essay, “Some Slavic Echoes in J.R.R. Tolkien’s Middle-earth” [3].

Kuzmenko’s essay therefore doesn’t break new ground, which unfortunately tends to undermine its mission of convincing us that Slavic elements in Tolkien’s writings deserve greater attention and investigation. Tolkien himself would seem to discourage the mission, having said that “Slavonic languages are for me almost in the same category [i.e., no aptitude]. I have had a go at many tongues in my time, but [...] the time I once spent on trying to learn Serbian and Russian have left me with no practical results, only a strong impression of the structure and word-aesthetic” [4].

There is a problem inherent in arguments like those of Orr and Kuzmenko; actually a couple of problems. For one, the mere existence of Slavic cognates to Germanic words we know were of interest to Tolkien says nothing about whether Tolkien was or was not aware of them, let alone whether he intended them to be detectable by readers (with one exception: “Variag”). For another, we often don’t know which cognates were the older ones, the Germanic of Slavic, though in some cases — as of, say, Gothic versus Old Church Slavonic — it’s pretty clearly apparent which is older. Third, even if a Slavic form is older and was borrowed into Germanic, we have no reason to assume either that Tolkien knew this, or that he cared if he did know. And if we are only interested in cognates, why not trot out a whole wealth of Sanskrit words? Why not? Because there is absolutely no evidence that Tolkien had Sanskrit in mind at any stage of the development of Middle-earth. One can make almost as definitive a rejection of the Balto-Slavic branch — but instead of absolutely no evidence, it’s rather simply very little. Certainly much less even than the Goidelic branch of the Celtic family, which Tolkien himself largely rejected.

For one example of something that has been made too much of, take Medwed, the original name for Beorn. Yes, this name is incontrovertibly Slavic. But why should this be cause for Slavists to rejoice? Tolkien rejected the name before publication of The Hobbit, replacing the Slavic name with a Germanic one. The fact of that rejection says something about Tolkien’s attitude toward the suitability of Slavic elements in Middle-earth, and should probably discourage the overenthusiastic search for others.

Note that I don’t mean to discourage all research in this area — far from it; I’m thinking only of the really overzealous, “no stone unturned” searches. As I indicated above, there is one very clear exception case, in Variag. This form, retained in the final published Lord of the Rings, is undeniably Slavic, from the Russian варяг, in turn from Old East Slavic (i.e., Old Russian, sometimes called Old Ukrainian) varęgŭ. The word is still in use today, actually. In Belarussian, as I understand it, вараг means both Varangian, but also (colloquially) any “tall, burly man”, and in the Arkhangelsk dialect of Russian, варяза is used to mean “a man from beyond the sea, a foreigner” (this is almost exactly, but coincidentally, analogous to the Semitic ferengi, which I have discussed before).

But aha! Despite the clearly Slavic word-shape Tolkien chose, its source is ultimately Germanic. Cognates include Old English wǽrgenga, Lombardic waregang, Old Frankish wargengus, cf. medieval Greek βαράγγως — all ultimately sourced to Old Norse væringi, váringi, from Proto- Norse *váringr, believed to derive from vár “vow, oath, pledge” + ganga “to go”, in the sense of foreigners pledged to service. [5] The word isn’t attested in Gothic, but if the Goths used the word, it would most likely have taken the form *wadjagagga, from wadi “pledge” + gagga “to go” (cf. the genuine faúragagga “a steward, lit. fore-goer”). The point is that although Tolkien chose a Slavic form for its presentation in the Middle-earth of the Third Age, the word, the very idea, of the Variags has its roots in Scandinavia, much as Beowulf has roots in Scandinavia but is seen through a distinctly English lens. And whatever (few) other Slavic echoes one might detect in Tolkien’s legendarium should, I expect, be regarded in the same way.


[1] Kuzmenko, Dmitry. “Slavic Echoes in the Works of J.R.R. Tolkien.” Literary Studies (National University of Kiev) 24 (2009): 217–221. All quotations are in English, machine-translated, then edited for sense by me. If you speak Ukrainian, please feel free to suggest corrections.

[2] Kuzmenko gives no source for the claim that Russian left an imprint on Quenya; however, he must have gotten this from Ivan Derzhanski’s entry, “Russian Language”, in Michael Drout’s Tolkien Encyclopedia (pp. 581–2). Kuzmenko includes Drout as an entirely generic entry in his bibliography, but does not explicitly cite it in the essay.

[3] Orr, Robert. “Some Slavic Echoes in J.R.R. Tolkien’s Middle-earth.” Germano-Slavica 8 (1994): 23–34. This is a solid essay, worth reading, but I think Kuzmenko may have relied rather too heavily on it. Kuzmenko’s assertion that “several images [...] at first glance have a Germanic origin, but on closer examination reveal Slavic parallels” is uncomfortably similar to Robert Orr’s “elements which at first appear to be simply taken from Germanic, but which on closer inspection appear to have various sorts of Slavic associations” (p. 23). Moreover, the three examples Kuzmenko gives — Mirkwood, Warg, and Smaug — are the same three Orr discusses, in the same order.

[4] Tolkien, J.R.R. The Letters of J.R.R. Tolkien. Ed. Humphrey Carpenter. Boston: Houghton Mifflin, 1981, p. 173 (#142).

[5] Blöndal, Sigfús. The Varangians of Byzantium. Trans, rev., and rewritten (!) by Benedikt S. Benedikz. Cambridge: Cambridge University Press, 1978. Reprinted 1981, 2007. pp. 4–6.

The World and Man’s essential place in it

Okay, I admit that sounds a bit overblown and vaguely philosophical — calling to mind subjective idealism, phenomenalism, the intellectual debates of Locke, Berkeley, and Hume, and so on. But while that’s an interesting subject in its own right (feel free to stop by Wikipedia on your way out), I had something else in mind. As usual, it all starts with words. So perhaps I might restate the title of this post as something more like: “world” and the essential place of “man” in its etymology.

If language really does define our perceptual limits (I won’t digress into a discussion of the theories of Sapir/Whorf), then it makes sense that we might find “man” in the “world” — human beings can’t really imagine a world without them in it, after all. I know I can’t! ;)

The word world derives from Old English weorold (interestingly, as you’ll soon appreciate, a feminine noun — I wonder whether this is the same grammatical process responsible for distinctions like the French le mort “a dead man” versus la morte “death, as an abstract concept”). [1] Skeat points out in his Etymological Dictionary that the word is clearly a composite form [2]. Cognate forms among the other Germanic languages include Old Norse (and Modern Icelandic) veröld, Old High German weralt (cf. Modern German Welt), Old Low Franconian uuerolt, and though not attested, I feel confident theorizing Gothic *waíralds [3].

Of what individual elements, then, might weorold be composed? Spenser thought it was war + old ; that is, the domain of an ever ongoing strife. So he says in the Faerie Queene — “But when the world woxe old, it woxe warre old / (Whereof it hight)” [4]. A clever guess, and one that seems apt enough given mankind’s perennial struggles, but wrong. Rather, the word comprises the two joined elements of man + age = “the age of man” = therefore, the world. Thus, OE wer “man” (cf. Modern English werewolf = “man-wolf”, and cf. Latin vir “man”) + eald “old, ancient, aged” (cf. eald-dóm “age”); and likewise, ON verr + öld; OHG wer + alt; OLF uuer + olt — all from Primitive Germanic *weraz + *alda.

Interesting, no? And it turns out that world is hardly the only word dependent on man. Consider Latin sæculum “lifetime, age of man” (from one of the oldest known Indo-European roots, meaning “to tie, bind”). As it developed, the Latin word retained both the senses of time, cf. Modern French siècle “century”; and world, cf. Modern English secular, meaning “worldly, mundane”. Speaking of mundane, which itself derives from the Latin word mundus “universe, world” (cf. French monde, Spanish mundo, Italian mondo), it’s tempting to look for the man in mundus as well, but that’s a misstep. The original meaning of mundus is “order”, as in a world or universe arranged neatly according to a divine plan. (The Greek κόσμος carries the same original sense and survives into Modern English in the words Cosmos, cosmic, and as a combining form in words such as cosmology.)

It is equally tempting to look for man in human, but the etymology of that word is a little different. Even so, we still can find a connection to the world in it. Turning back to the familiar Latin source, the adjective hūmānus “pertaining to man”, from the noun homo “man”, we find something surprising: homo derives from humus “the ground, the earth, the soil” from which, according to the Christian tradition, the first Man came [5]. The word humus has survived into Modern English as well, referring to fertile, organic topsoil suitable for growing things. Where did the Romans get this idea? From the Hebrew Bible, one would surmise. There, the first Man took his name, Adam (Hebrew אדם), from the substance out of which God made him, the earth (Hebrew אֲדָמָה /adama/). Genesis 2:7 [6] may therefore be one of the oldest statements of etymology we know!

How much of this connection between man and the earth is due to our own anthropocentric view of the universe? All of it, really; Man was, and is, the central actor on the stage of the world (to paraphrase Shakespeare). After all, without Man, who would there be to speak the word world — or anything else, for that matter? I suppose our glossopoeic nature gives us license to name, even to define, the world with our own distinctly human nomenclature. If we one day should find that chimpanzees, dolphins, or bees have genuine languages of their own, then who knows? We may have to rethink our whole Weltanschauung. :)



[1] Does it go without saying that there are several different dialectal variations of the word, as of most Old English words? It should. “Old English” usually corresponds to the West Saxon dialect, simply because of the extent of the surviving lexis, though Modern English, in fact, is closer to the Mercian dialect. At any rate, other attested forms include weoruld, weorld, woruld, worold, etc.

[2] Skeat, Walter William. An Etymological Dictionary of the English Language. 2nd ed. Oxford: The Clarendon Press, 1893, p. 718.

[3] Built from the same components as the other cognate forms; in this case, the attested Gothic waír “man” + alds “age, life” (see Wright, Joseph. A Primer of the Gothic Language. Oxford: The Clarendon Press, 1892, pp. 238, 212, for these words. Putting them together, however right or wrong, was me.) There is also an attested word with a similar composite meaning in Gothic: manaséþs “mankind, multitude, world” (lit. “man-seed”).

[4] Book IV, Canto VIII, ll. 276–7.

[5] Valpy, F.E.J. An Etymological Dictionary of the Latin Language. London: Baldwin, Longman, and Whittaker, 1828, p. 188. Note that other mythologies explain the origin of Man differently; for example, in the Old Norse tradition, the first Man and Woman were Ask and Embla, the Ash and Elm tree, respectively.

[6] In the King James translation, “And the Lord God formed man [Adam] of the dust of the ground, and breathed into his nostrils the breath of life; and man became a living soul,” emphasis added. (The original Hebrew, וייצר יהוה אלהים את האדם עפר מן האדמה ויפח באפיו נשמת חיים ויהי האדם לנפש חיה)

“New Learning and New Ignorance”

You know what I love? When you’ve made an academic argument based on the best evidence you were aware of at the time, and then you find new evidence, long afterward, which backs up your theory! It also points out just how much “evidence” there is out there — both for and against any given proposition. Finding it, sifting through it, and evaluating its merit, that’s the great challenge.

A little over a year ago, I presented a paper at Mythcon 37 in which I attempted to trace the most likely sources for Tolkien’s well-known, but usually overlooked, literary collocation “Circles of the World.” The three I discussed were the Old Norse Heimskringla, the Biblia Vulgata, and the famous mappa mundi at Hereford Cathedral in the West Midlands. The two philological analogues here are the Old Norse kringla heimsins “circle of the world” and the Latin orbis terrarum “circle of the worlds”, and though I think I was successful in plausibly connecting both to Tolkien, I didn’t have a good transition from the one to the other — apart from the rather obvious fact that they mean the same thing. Well, thanks to serendipity — and to a diet of abstruse reading :) — I’ve now discovered a single source, one that Tolkien definitely knew, that connects these two precedents (as well as other conspicuously Tolkienian concepts) in the space of just a few pages. Curious?

I found it in Volume II of Jacob Grimm’s Deutsche Mythologie (1883–1888). This is in the James Steven Stallybrass translation, which he retitled Teutonic Mythology. I was reading Chapter XXV Time and World, when I came across this (with apologies for the choppy editing):

The ON. heimr is mundus, domus, and akin to himinn, himil (p. 698), as mundus also is applied to world and sky; heimskrîngla, orbis terrarum. Ulphilas renders οικουμένη [...] by [Gothic] midjungards; to this correspond the AS. middangeard [...]; the OHG. mittingart [...]; the OS. middilgard; the ON. miðgarðr [...]; and even a Swed. folksong [...] has retained medjegård. [1]

This is quite something, to see my two philological sources separated by no more than a comma! And compellingly followed by a laundry list of precursors for “Middle-earth”, one of which (the Greek οικουμένη) Tolkien mentioned explicitly in a letter [2].

As Grimm continues, more tantalizingly Tolkienian references appear. These include linking orbis terrarum to the idea of the “sea-girt world,” with cognates and sources aplenty, most notably the Old Norse miðgarðs ormr, the “serpent [...] coiled round the earth’s circumference, [...] evidently the ocean.” This, too, finds a point of connection with my paper, in an argument suggesting that the Hereford mappa mundi, also depicted as girded round by water, may find an echo in Tolkien’s Ekkaia, the Encircling Sea in his fictive geography.

Then, at the bottom of the page, Grimm surprisingly produces the Finnish ilma “air, sky”, which many of you will recognize for Tolkien’s use of the same word in Quenya, with an obviously related meaning (“starlight”); as well as in Ilmarin (“Mansion of the High Airs”), the palace of Manwë and Varda atop Taniquetil; and even Ilmarë, the handmaid to Varda. Tolkien, we’re pretty sure, would have gotten this word from the Finnish Kalevala, but he evidently saw it here too.

And even more, as Grimm continues over the next two pages, we see mention of the Völuspá, a self-acknowledged source for Tolkien; Yggdrasil, the World Ash, connected by many scholars to Tolkien’s own metaphoric and mythopoeic use of the Tree imago; and specific mention of the Old Norse dwarves “Dâinn [sic] and Dvalinn,” both names memorably recycled by Tolkien. All of this, from start to finish, in just three consecutive pages of a text Tolkien knew — and knew well. It’s therefore no great stretch to see Deutsche Mythologie as having gone into the same leaf-mould that produced Tolkien’s “Circles of the World.”

I love that! Now, to incorporate this into a revision of my paper!

[1] Grimm, Jacob. Teutonic Mythology, Volume II. Trans. James Steven Stallybrass. Dover Phoenix Edition. Mineola (NY): Dover, 2004, p. 794.

[2] Tolkien, J.R.R. The Letters of J.R.R. Tolkien. Ed. Humphrey Carpenter. Boston: Houghton Mifflin, 1981, p.239 (#183).

Writhe, writhen, wroth … wraith? And ring?

In the absence of any direct depiction of Sauron — who works much better as a presence felt and feared, but never seen — evil in Middle-earth is most effectively embodied in the Nazgûl, the Ringwraiths. Tolkien’s Ringwraiths are usually seen enrobed in black, faces unseen, and in fact invisible, sowing the seeds of despair wherever they go. The image has been taken up in fantasy literature in many forms since — e.g., think about Grimnir in Alan Garner’s Weirdstone of Brisingamen, or consider the Dementors in J.K. Rowling’s Harry Potter novels.

But Tolkien didn’t invent this word. It was used much earlier by Sir Walter Scott (where Tolkien may have encountered it), among many other instances. So what is a wraith, actually? What does this rather curious word mean? It’s usually used for some sort of ghost, apparition, or premonitory omen of death (something like the Grim in J.K. Rowling) as here.

But regardless of how it’s used, what does it mean? For the best sense of that, we have to turn to its etymology. Yet as you can see at Encarta — “[Early 16th century. Origin ?]” — its etymology is not terribly obvious. So let’s look a bit further back. In his influential Etymological Dictionary of the English Language, Walter Skeat can only conjecture (see the illustration at the top of this post). One suggestion connects it to Icelandic vörðr “warden, guardian” [1].

What else do we have? How about John Trotter Brockett’s Glossary of North Country Words? And no, I’m not making up Brockett’s middle name — something that would have tickled Tolkien’s fancy, I imagine! Brockett connects wraith to a northern dialectal word, waff, meaning “an apparition in the exact resemblance of a person, supposed to be seen just before or soon after death.” He guesses the word may be linked to waft, no doubt envisioning the airy apparition wafting along in a gentle breeze on the moors. [2]

Poking around in various other etymological dictionaries, we find other suggestions, some good, some not so good — including Old English weard, ward “warden, guardian” (cf. Skeat’s vörðr); Old Gaelic breith (aspirated form, bhreith) “doom, judgment”; and even Modern English wrath, wroth (from Old English wráð), suggesting a wraith might be a particularly wrathful spirit. The OED traces the word all the way back to 1513, to a translation of the Aeneid made by the Scottish poet, Gavin Douglas; consequently, they assume its origin to be Scottish. [3]

Any or all of these are possible, but Tolkien had a different etymology in mind. He thought wraith might have been a form of the verb writhe “to twist or struggle, as in pain”, with the archaic past participle, writhen (< Middle English writhen < Old English wríðen “to twist, torture”). The idea being that a wraith was something twisted, contorted, bent out of its proper original shape — in the case of the Nazgûl, by evil. [4]

This would, in fact, cousin the word to wrath, wroth, which has a literal meaning closer to “twisted by rage” than is usually remembered. Its Old English form was wráð, derived from Primitive Germanic *wraithaz. Now that's beginning to look familiar!

Other interesting cognates include Gothic wraiqs “curved, winding, twisting”, and perhaps wraka “persecution, punishing pursuit; Old Frisian wreth “evil”; Old Saxon wred, Middle Dutch wret “cruel”; Old High German reid, Old Norse reiðr “wroth”; Old Norse ríða “to twist, knit, wind” — all originally related to the Proto-Indo-European root *wreit “to turn, wind”. Another descendent of this root is Modern English wreath, from Old English wriða “band (i.e., that which is wound around)” — sounds rather like a ring, no? Indeed, in the Old English corpus, wríða is sometimes used for “ring” (as in the Homilies).

From “wraith” to “ring” in only, what, a dozen complicated steps? :) And I'm not even remotely finished with this topic! Stay tuned for more to come.


[1] Skeat, Walter W. An Etymological Dictionary of the English Language. 2nd ed. Oxford: Clarendon, 1893, p. 720.

[2] Brockett, John Trotter. A Glossary of North Country Words with Their Etymology and Affinity to Other Languages and Occasional Notices of Local Customs and Popular Superstitions. 3rd ed. Newcastle Upon Tyne: Emerson, Charnley, Bigg Market; and Simpkin, Marshall, & Co., 1846, pp. 200-1, 202.

[3] Gilliver, Peter, Jeremy Marshall, and Edmund Weiner. The Ring of Words: Tolkien and The Oxford English Dictionary. Oxford: Oxford University Press, 2006, pp. 223-4.

[4] Shippey, Tom. The Road to Middle-earth: How J.R.R. Tolkien Created a New Mythology. Rev. and expanded ed. Boston: Houghton Mifflin, 2003, pp. 148-50.

Wulfs, waúrms, and *wargs — oh my!

Unpacking some boxes from storage a couple of weekends ago, I unearthed a long-forgotten gem: Brian Regan’s Dictionary of the Biblical Gothic Language. Out of print, unavailable from Amazon or BookFinder, I’m glad I never got rid of this. It originally belonged to my oldest friend, Gary Schmidt, but he passed it on to me some time in the middle 1980’s.

In his introduction, the author calls this “the first complete Gothic-English dictionary in eight decades” [1]; so I’m torn between being relieved that the complete lexicon of attested forms is short enough for convenient use (about 200 pages of fairly large type) and the sad realization that so many, many words of 4th Century Gothic are now lost forever.

To bring this around to Tolkien (more or less inevitable with me), it’s pretty well known that he developed a soft spot for Gothic after first discovering it in his mentor Joe Wright’s Primer of the Gothic Language. He wrote a pretty well-known poem in it (“Bagme Bloma” — Gothic for “Flower of the Trees”), included in the rare collection, Songs for the Philologists. He also incorporated elements of Gothic names into the ancestors of the Rohirrim as part of the backdrop for The Lord of the Rings. And Tolkien was known for deploying his extensive understanding of comparative Germanic phonology in an effort to extrapolate the forms of unattested Gothic words. (Such words are generally preceded with an asterisk, as in *wargs in the title of this post.) There’s even an extended piece of Gothic in one of Tolkien’s letters — a facsimile of an inscription he made in a book he owned in 1910 (the image at the upper right is a part of that inscription). [2] My friend Gary once adapted this same inscription on the flyleaf of a Dutch-English dictionary he gifted me back in the early 1980's; sadly, I'm not sure I have it any longer. Maybe one of these days, I'll get a letter inquiring about it, as Tolkien did in 1965.

Fellow admirers of Tolkien will probably recognize the Gothic words in the title of this post, too, even without any special training. For those who don’t, they are: “wolf, worm (i.e., dragon), outlaw (> warg)”. They’re actually in the nominative singular form, but to speakers of English they look more like plural forms this way, so I left them. For anyone wanting to take a stab at Gothic (but who might be too intimidated by Joe Wright), David Salo (known for his work on Tolkien’s invented languages, especially Sindarin) has put together a simple — and incomplete, it must be added — outline of Gothic grammar.

(Okay, I'm betting this post will have scared off the few readers I had! Anyone who’s still reading, drop me a comment! ;)

[1] Regan, Brian T. Dictionary of the Biblical Gothic Language. Phoenix: Wellspring, 1974, p. vii.
[2] Tolkien, J.R.R. The Letters of J.R.R. Tolkien. Ed. Humphrey Carpenter. Boston: Houghton Mifflin, 1981, pp. 356-8.