Éala Éarendel: A study in names

A meditation on words, slightly late, but suited for Eastertide. Any howling errors herein are wholly my own; though I reserve the right to be an intellectual coward, and blame them on my recent concussion.


There is no such thing as an expert on language. There are experts on individual languages, and experts on different aspects of language as a phenomenon; but the field of language as a whole is, and always has been, far too large for anyone to adequately survey in a human lifetime. Tolkien came as near it as almost anyone: he was intimately familiar with the whole 1,500-year history of English, plus Old Norse, Latin, and classical Greek, and had a firm working knowledge of German, French, Spanish, Welsh, Irish, Hebrew, and several other languages, including the latest reconstructions of Proto-Indo-European. Yet he wrote, with perfect sincerity, to Fr. Robert Murray: ‘I am in no ordinary sense a “linguist”’. He understood better than most professional linguists the internal workings of language, but he also had a sound knowledge of his own limitations.

It may be unfair to compare Tolkien with Noam Chomsky, who does unabashedly call himself a linguist, and is often regarded by his younger colleagues in the field as the linguist. Unfair, but for my present purpose, necessary. Chomsky does not show any signs of great familiarity with any language but English. He attempts to lay down ‘universal’ rules of grammar, but his universals, when closely examined, tend to be disturbingly parochial.

For Chomsky, the whole and sufficient business of linguistics is generative grammar – the rules for assembling words into sentences. Where the words themselves come from does not much interest him, and he has famously accused etymologists and philologists en bloc of asking ‘the wrong questions’. Yet having so restricted his field of inquiry, he restricts himself still further by making grammar itself largely a matter of word-order and the assembly of phrases. This leaves out, for instance, the so-called polysynthetic languages, which can cram a whole sentence’s worth of meaning into a single word, by bolting long strings of prefixes or suffixes onto a single root. It even leaves out languages like Latin and Greek, where the order of words in a sentence is almost arbitrary, and the grammatical role of each word is determined by its inflections.

I will offer a single example before moving on. Linguists of Chomsky’s school talk of ‘pro-drop’ languages, meaning those languages in which one can ‘drop’ the subject of a sentence when it happens to be a pronoun. This, in fact, is a singularly inapt description of what happens. In languages like Latin and Greek, or (to take the example best known to me) Spanish, the pronoun is not dropped; it is fused with the verb. When Horace says Odi vulgus profanum et arceo, he is not omitting the pronoun ego from the sentence. The Latin for ‘I hate’ is not ego odi, it is simply odi; ‘I ward off’ is not ego arceo, but simply arceo. The history of sound-changes has obscured the first-personal ending of odi, but in arceo we see it standing plain and proud in its primitive Indo-European glory: it is O. We find the same ending in Greek, or in Spanish, when I say, for instance, Creo que Chomsky no siente vergüenza. Creo – ‘I believe’. If I alter that to yo creo, according to the ‘pro-drop’ people, I am simply restoring the ‘natural’ form of the sentence; but that, in Spanish, is not what I am actually doing. Adding the pronoun yo is a duplication for emphasis. Yo creo translates more accurately: ‘I myself believe’, or ‘Me, I believe’. But this is the sort of thing that Chomskyites miss, because they take the peculiarities of English grammar as normative.

I do not, of course, pretend to be an expert on language; nor in any particular language – not even English, which is the cussedest and most complex language on the face of the earth, and cannot be thoroughly understood without a knowledge of a dozen others. I know just about enough linguistics to tell a real linguist from a poseur; to tell, for instance, that if Tolkien is not a linguist, Chomsky does not even begin to be one. The good Chomskyites who taught me their version of linguistics were as parochial as their patron: I could have got my degree from them without any knowledge of a language other than English. There is a second-language requirement for the M.A. program, but only because German is still the lingua franca of linguistics, and it is necessary to be able to read the relevant academic papers. In fact, it would appear, these modern linguists have never even heard of the dictum that a person who knows only one language does not really know even that.

My point, so far, is that language is a minefield, and even the most learned of men may happen to step on a mine. When amateurs wander in, they have a way of blowing themselves up in a matter of seconds. It is particularly good fun to see one amateur go into the minefield to try to guide another. In The Age of Faith, Will Durant (who was a fairish popular historian but no linguist) discusses the Etymologiae of St. Isidore of Seville, who was, sadly, the nearest thing that Visigothic Spain possessed to a learned man. Isidore’s book, despite the name, is a one-man encyclopaedia, a compendium of the important bits out of dozens of classical authors, not always fully digested or understood. It remains important today chiefly because so many of his original sources have been lost, and would be forgotten entirely if Isidore had not preserved fragments of them. Durant acknowledges this, but he has a low opinion of Isidore’s scholarship, calling it (among other things) ‘a farrago of weird etymologies’. And almost at once, he steps on a mine.

The trouble about etymology is that words that resemble one another are often unrelated, but words that really are related are often very different. The words isle and island, so obviously alike in sound and meaning, are completely unrelated; whereas five and punch (the drink, not the thing you do with a fist), which seem completely unalike, actually derive from the same Indo-European root. Classical etymologists liked to make up just-so stories about the origin of words, superficially plausible, but actually nonsensical; Isidore was firmly in this tradition. The catch is that they were not necessarily wrong; and it takes a real etymologist to tell the difference.

So Durant ridicules Isidore for thinking that the knees are called genua in Latin because ‘in the foetus, they lie opposite the cheeks (genae)’. So far, so good: the similarity of sound is accidental. Gena comes from the same root as Greek γένυς and English chin, and seems originally to have referred to the jaw. Genu comes from the same root as Greek γόνυ and English knee. (From γόνυ, by the way, comes γωνία ‘angle’, from which we get words like polygon. Etymologically speaking, a hexagon is a figure with six knees.)

But while Durant’s right foot lands safely, his left foot is busy stepping on a mine. He also laughs at Isidore for saying that man ‘is called homo because God made him from the earth (humus)’. In fact, homo and humus really do go back to the same root, which the Indo-Europeanists reconstruct as *dʰéǵʰōm, meaning ‘earth’. From this, we are assured, comes *dʰǵʰémō, ‘earthling’ – though we cannot now tell whether that meant ‘one who lives on the earth’ or ‘one who is made out of the earth’. At any rate, it is not an unusual connection for a language to make. The Hebrew word אָדָם (adam), which everybody knows as the name of the first man, meant both ‘man’ and ‘soil’. In fact, Durant is making fun of Isidore for being exactly right.

If there is a lesson in all this, it is that a folk etymologist is like a stopped clock. If anything, this underrates the clock; for there are not many folk etymologists that manage to be right twice a day, and stopped clocks don’t usually make a nuisance of themselves by loudly announcing what time it is. The folk etymologists are often assiduous about setting the record wrong. At various times they insisted on changing abominable to abhominable, and foxglove to folks’-glove, and coconut to cocoanut – and were mistaken each time. (And as the immortal H. W. Fowler said: ‘Welsh rabbit is amusing and right. Welsh rarebit is stupid and wrong.’) They also, in the 16th century or thereabouts, stuck a silent S in the middle of the Anglo-Norman word ile,  just because the Latin insula has an S in it; then they stuck the same S into iland, supposing that it was related to i(s)le. They even buffaloed the whole English-speaking world into abandoning the spelling delite, which was both etymologically and phonetically correct, for the ridiculous delight.

Nowadays, we have a rash of folk etymologists who are also village atheists. An early sign of spring, in most years, is that these people turn up all over the Internet, solemnly informing us that Easter is actually a pagan festival which the wicked Christians took over and ruined. In support of this idea they quote the Venerable Bede, who claimed that Éostre or Éastre was the name of a pagan Anglo-Saxon goddess connected with springitme. Latterly they have begun to claim that Easter (or Éostre) is identical with Ishtar, and that the holiday originated as the feast of a Near Eastern fertility goddess.

All this – it should be, but is not, needless to say – is bosh. To begin with, while the name ‘Easter’ is taken from pre-Christian English, the holiday is not. That was brought over from the Continent by the early Christian missionaries, who called it (in both Latin and Greek) Pascha. This name, appropriately enough, comes from Hebrew Pesach (פֶּסַח), which in English is called Passover. (It has also, in a relatively harmless example of folk etymology, been linked with the Greek word πάσχω ‘I suffer’.) The Romance languages call the holiday by names derived from Pascha: Pascua, Pasqua, Pâques, etc. It becomes Pasg in Welsh, påsk in Swedish, Pasen in Dutch. Only English (Easter) and German (Ostern) use names taken from a pagan Germanic term.

Now what about Éostre herself? The evidence for the existence of this goddess is rather sparse. Apart from Bede, we have Jacob Grimm drawing comparisons (nearly all based in comparative linguistics) with goddess-figures from various Indo-European cultures with related names. But Grimm had a tendency to claim too much. He liked to believe that if words in related languages could be traced back to a common root, the meanings of the words must be of equal antiquity. This is not necessarily the case. Our word beech goes back to a PIE root, reconstructed as *bʰeh₂ǵos, but that word did not necessarily refer to beech-trees. It also gave rise to the Greek φηγός, which means ‘oak’, and Russian buzina, which means ‘elder-tree’.

The endless arguments about the original homeland of the Indo-European people turn largely on problems like this. There are PIE words that appear to mean not only ‘beech’, but ‘salmon’, ‘lake’, ‘horse’, ‘snow’, and so on; and there are other things, like salt seas and oceans, for which there appears to have been no word in the language. If you go looking for a place on the map that had all the things for which there were PIE words, and did not have the things for which there were no words, you will find yourself on a wild-goose chase. What you have to remember is that the words themselves have changed their meanings. To figure out exactly what they meant 5,000 years ago, and where all the things they originally referred to could then be found, is probably an impossible task.

However, there are clues. There does not seem to be any solid evidence of an actual cult of Éostre; the alleged goddess is not even mentioned in any surviving source except Bede. If you go back in time to Proto-Germanic Austro, or to PIE *H₂ewsṓs (or Ausos), you will find still less to go on. The Greek equivalent, Eos, and the Latin Aurora, usually occur not as proper names, but as ordinary words meaning ‘dawn’. It is true that Homer is constantly referring to Eos, ‘the rosy-fingered dawn’, as a goddess; but this may be no more than a poetic personification.

No more; and no less. We can take another clue from Owen Barfield’s Poetic Diction. In ancient times, and particularly in the ages before the Greeks took up philosophy, there was no hard distinction between natural phenomena and persons. It was pre-eminently the age of what Martin Buber calls the ‘I-Thou’ relationship. Anything in nature that seemed to have agency – anything that did things – was naturally considered to have a spirit, and to be something resembling a person. This is not to say that such things were always anthropomorphized. The Romans, for instance, thought of the powers of nature as numina, divine spirits that might have very little in common with the souls of men. They could, in principle, be communicated with, and one could make bargains with them; some of them, it seemed, could be appeased by sacrifices and other rituals. But they did not necessarily have thoughts or emotions as we understand them. The pagan divinities of the Celtic, Germanic, and Slavic peoples were not necessarily much more human than this. It was only the Greeks, through their prolific invention of myths and poems, who made a definite and unanimous decision to represent all their gods as superhuman beings in human form, using human language, and having human thoughts and feelings. (They also, in the process, made their gods seem monstrous and even ridiculous. A man who carried on with everything female under the sun, as Zeus was supposed to do, would be downright disgusting – even by Greek standards. Euripides and Plato, among others, were sadly aware of this.)

Eos or Aurora, then, was the dawn, the brightening of the sky before sunrise; but equally, the numen of dawn, the spirit of the early morning. Nobody seems to have sacrificed to the dawn, though they often sacrificed to other gods at dawn. Hesiod fitted Eos into his elaborate genealogy of the gods, making her the daughter of the Titans Hyperion and Theia, and the sister of Helios and Selene (the Sun and Moon). On the other hand, Diodorus Siculus, writing at the very end of the development of Greek myth, says:

Of Hyperion we are told that he was the first to understand, by diligent attention and observation, the movement of both the sun and the moon and the other stars, and the seasons as well, in that they are caused by these bodies, and to make these facts known to others; and that for this reason he was called the father of these bodies, since he had begotten, so to speak, the speculation about them and their nature.

This, of course, is partly the intellectualizing Hellenistic tradition at work; but it reminds us just how changeable the Greek myths were. Aphrodite, according to Homer, was the daughter of Zeus and Dione; but Hesiod made her the daughter of Uranus, formed from the foam (ἀφρός) where his genitals fell into the sea after Cronus cut them off. Those who see Éostre–Eos–Ausos in every closet would have us believe that Aphrodite ‘is’ Eos as well; but for all the stories that the Greeks have told about Aphrodite’s origins, that is one that has not come down to us. None of the genealogies make her a Titaness, or a close relation of Helios and Selene.

I am afraid that in this case, the most naive and simple folk etymology is the one that will bring us closest to the truth. The obvious thing is to think that ‘Easter’ and ‘East’ are related; and this, in fact, is true. East is defined, originally, as the direction of sunrise; compare Latin oriens, which means both ‘east’ and ‘rising’. By an obvious analogy, sunrise is associated with springtime: the spring is the morning of the year, the time of increasing light. In the evening, as we still say in English, the sun westers; it approaches the horizon in the West, and disappears beneath it. There is no verb ‘to easter’, since the sun is never seen to approach the horizon in the East; but it does so all the same, out of sight, hidden by the bulk of the earth. If we had such an expression, we should say that dawn is the brightness in the sky caused by the eastering sun; and that springtime is the eastering of the year. And in fact it is this last expression that the Anglo-Saxons used, almost exactly, in the phrase éostre tid — Eastertide.

Where does this leave Ishtar? She was the Akkadian and Babylonian goddess of love and war (in which we are told that all is fair), as well as sex and fertility. The Greeks identified her as a love-goddess with Aphrodite, and as a war-goddess with Artemis; but the Greeks never supposed that Aphrodite and Artemis were the same person. Ishtar was identified with the evening star before Babylonian astronomers figured out that the morning and evening stars were one and the same planet. It is because the Greeks partially identified Ishtar with Aphrodite (and the Romans identified Aphrodite with Venus) that we now call that planet Venus.

The folk etymologists and some neopagan syncretists, the ones who conflate Aphrodite together with Eos, would have us believe that both of these Greek goddesses are also identical with Ishtar; but Ishtar was not, by any evidence available to us, a goddess of the dawn. The idea that ‘Ishtar’ and ‘Easter’ are related words is a silly notion arising from a misunderstanding of the various pagan syncretisms involved, and from a completely fortuitous resemblance of sound. There is in fact no discoverable connection between the name Ishtar and any Indo-European word. The same, of course, goes for Ashtart, the Syrian–Phoenician version of the name, and for Astarte, which is how the Greeks rendered it into the sounds of their own language.

What, then, of the morning star? This, it would seem, was not identified with Eos by the Greeks; probably because the dawn occurs every day, whether Venus is visible in the sky or not, and at the most productive period of Greek myth-making, they had not yet acquired the astronomical knowledge of Babylon. It was the Babylonians, so far as I know, who first gave the seven traditional planets (including the sun and moon) the names of their gods; and when the Greeks did take up astronomy, they gave the planets the names of roughly equivalent gods. The Romans repeated the process. So Marduk became Zeus, which became Jupiter; Ishtar became Aphrodite, which became Venus; and so on. We can trace this process of substitution one step further, in the modern English days of the week. Tuesday, Wednesday, Thursday, and Friday are named for Tiu (Mars), Woden (Mercury), Thor (Jupiter), and Freya (Venus) – these being the rough and ready equivalents between the classical and Germanic gods. Freya is not Éostre, any more than Aphrodite is Eos (or Artemis). The identification did not run in that direction.

However, the Germanic peoples had their own myths about the morning star, unconnected with Freya or her sisters from the South. According to Snorri Sturluson’s Prose Edda, Aurvandil the Valiant was Thor’s companion on a journey. At one point, Thor carried Aurvandil out of Jötunheim in a basket. One of Aurvandil’s toes stuck out of the basket, and in that place of eternal cold and frost, it froze solid. Then Thor broke off the toe and threw it up into the heavens, where it became the star known as Aurvandil’s Toe. This has traditionally been identified with the morning star.

Aurvandil, as it happens, is a kind of linguistic relative of Éostre. The Primitive Germanic version of the name has been reconstructed as Auziwandilaz, the ‘Bright Wanderer’ or ‘Dawn Wanderer’. Auzi is another form of the word that appears in East and Éostre. It is this connection between Aurvandil and Éostre that makes me reasonably certain that the morning star, and not some other heavenly body, is Aurvandil’s Toe. And it is the separate existence of the two that tells against any connection in the deep prehistory of mythology with Aphrodite or Ishtar. The Bright Wanderer is not the goddess Dawn; indeed, the Bright Wanderer is not a goddess at all, but a legendary hero’s frozen big toe.

The Old English form of Aurvandil is Éarendel, and the morning star came to be known by that name in England (without mention of the toe, it would seem). After the English converted to Christianity, many of their tales and legends were, so to speak, baptized: the people were too fond of them to abandon them wholesale, but they found ways of accommodating them to the Christian view of the world. So, while the astrologers were busy identifying pagan Venus with pagan Freya, an unknown poet made an ‘identification’ in a different direction – attaching the Germanic myth to a Christian story. In the Old English cycle of poems called Crist, Éarendel is identified with John the Baptist. Just as the morning star rises before the sun and announces its coming, John arises before Jesus and announces that the Messiah is at hand. Hence the famous lines:

Éala éarendel engla beorhtast
ofer middangeard monnum sended
and sodfasta sunnan leoma,
tohrt ofer tunglas þu tida gehvane
of sylfum þe symle inlihtes.

‘Hail, Éarendel, brightest of angels, over Middle-earth sent to men, and true radiance of the Sun bright above the stars, every season thou of thyself ever illuminest.’

Crist may have been written about the year 800; the Venerable Bede died in 735. By this time, Éostre had firmly taken its place as the name of the Christian holyday, and the pagan goddess (if she was ever more than a personification of daybreak) had been largely forgotten. But the identification was still there, available for poets to use. The Crist-poet (or poets) must have known of it, and made brilliant use of it. Éarendel, the forerunner of Éostre, the dawn, is taken up into Christianity to represent John, the forerunner of Christ – whose passion and resurrection, the centre of his story, are celebrated at Easter, the dawn-tide of the year. It was too good a match for a poet not to use; as good as Vergil’s Fourth Eclogue.

And so we come back to Tolkien, who was (in his own terms) no linguist, but knew pretty nearly everything that anybody still knows about Old English. While still a very young man (just a hundred years ago last Thursday, give or take) he read Crist and was thunderstruck by the lines about Éarendel. Here was a bridge between the pagan myths and legends that he loved and the Christian religion in which he devoutly believed; and both were better for it – the Christian world made more colourful and imaginative, the pagan suffused with hints of meaning and hope. Long ages ago, but probably still after the writing of Crist, a supreme English poet had taken a hint, maybe, from Crist, and written an epic about the old pagan days, informed and illuminated with foreshadowings of Christian grace; we know it as Beowulf. Young Tolkien took his cue from both. Henceforth he would earn his bread by teaching the old poems and the old languages in which they were written, but the work of his heart would be to write new works of the same kind and flavour. He, too, would build a bridge between pagan and Christian, a mythology for England, informed and sanctified by the knowledge of the true God. As a philologist, it would be Tolkien’s job to resolve cruxes, conundrums within a text or contradictions between texts. As a storyteller and poet, he would strive with the conundrum of human longing, the contradiction between pagan Fairyland and Christian Heaven; he would seek all his life to resolve the crux. For crux, after all, is just the Latin word for the Cross.


  1. Sherwood says

    That was fascinating!

  2. That was absolutely brilliant, and very illuminating. I loved how it all tied in together, eventually leading us back to Tolkien.

  3. I just love reading your essays. I know the subject matter, too, and you’re always spot on.

  4. The context of Tolkien’s letter makes it clear what he meant:

    I love music, but have no aptitude for it; and the efforts spent on trying to teach me the fiddle in youth, have left me only with a feeling of awe in the presence of fiddlers. Slavonic languages are for me almost in the same category. I have had a go at many tongues in my time, but I am in no ordinary sense a ‘linguist’; and 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.

    That is, he was not a person who could use many languages, which was the original sense of linguist (the first OED1 quotation is 1593); the sense ‘philologist’ (which is the word Tolkien used of himself) is not unambiguously found until 1922, when Jespersen used it. The first three quotations the OED gives for it, below, strike me as ambiguous. But a linguist in that sense he certainly was, and so is Chomsky; if their concerns are different, that is only natural in the growth of the field from singularity to diversity.

    1641 J. Wilkins Mercury iii. 23 Many of the other [words]..are of such secret sence, as I thinke no linguist can discover.

    1695 J. Edwards Disc. conc. Old & New-Test. III. i. 3 Here Linguists and Philologists may find that which is to be found no where else.

    1749 D. Hartley Observ. Man i. iii. §1. 320 A Light in which Grammarians and Linguists alone consider Words.

  5. > “Yo creo translates more accurately: ‘I myself believe’, or ‘Me, I believe’. ”

    We have a mild analogy to this in English in the rare cases where separate pronouns are semantically optional. “I cook, wash up and do the laundry” emphasises the actions. “I cook, I wash up and I do the laundry” emphasises the doer (“it’s me who does these things”).

    • “whereas five and punch (the drink, not the thing you do with a fist), which seem completely unalike, actually derive from the same Indo-European root”
      Likewise — and this did surprise me, but I checked it out — “feuding” and “feuding” are not cognate either.

      • Sorry, “feudal” and “feuding” = not related.
        An understandable error to assume that mediaeval knights and their lords were bound by “feudal” loyalty that required them to fight for each other when “feuds” arose. But no.

Speak Your Mind