Sunday, October 9, 2011

The Poros and the Bosphorus

Through the hot, seldom-traveled plain of southeastern Gondor runs an overlooked river, the Poros, southernmost tributary of the great Anduin. Running more or less east to west, it forms a natural boundary between the furthest reaches of Gondor and lands under the sway of Harad to the south. The Harad Road fords the river at the Crossings of Poros, continuing north through Ithilien to the Crossroads and still further to the Morannon, the Black Gate of Mordor.

In The Lord of the Rings, the Poros is barely mentioned. Apart from its proper place on the maps, it figures only in the appendices as a site of frontier skirmishes between Gondor and the Haradrim (Appendix A.I.iv; and see Appendix B at TA 2885). With so few references, why should this far-flung river be of any interest to anybody? Well, it’s the name that attracted my attention. In the context of Middle-earth and its languages, we don’t know what it means — and that is pretty rare.

Such puzzles always pique my curiosity, and I think I have an answer. Having a look through the materials available to me, and performing some moderately thorough (though not exhaustive) searches of the Internet, I don’t come across anyone with the same theory I am about to share. If anyone has seen this, please let me know. Anyway, here goes.

Tolkien doesn’t discuss the name in “The Rivers and Beacon Hills of Gondor” (Vinyar Tengwar 42); it seems not to be glossed in “Words, Phrases & Passages in Various Tongues in The Lord of the Rings” (Parma Eldalamberon 17); it’s not in the “Nomenclature” Tolkien prepared for translators; nor is it in the Eldarin Etymologies. It’s really a bit of a mystery. As a result, guesses as to the meaning of this name are just that — guesses. The common element in most of these guesses is Sindarin ros “foam, spray”, but the first element is pretty much totally unknown. Eldarin roots with similar sound silhouettes seem to be red herrings (“flour”? “north”?). Jim Allan once suggested that it might be the same element in the equally rare (and also appendiceal) name, Araphor (= aran + por), but this doesn’t help much since we still have no idea what the element por is supposed to mean. And that’s assuming the name is Sindarin at all. A welter of names in the south of Gondor are said to be of pre-Númenórean origin and not Eldarin. The fact is, we just don’t know.

Here’s my theory, something I’ve been meaning to share with you for a long, long time. I can’t help wondering whether the name might have a primary world etymology. After all, it looks like a standard form Greek noun of the second declension, doesn’t it? In fact, there is such a word. Ancient Greek attests πόρος, matching Tolkien’s spelling exactly, and what is more, its meaning is highly suggestive. Of several connotations and uses, there are these in particular: (1) “a means of passing a river, a ford or ferry”, and (2) “a narrow sea, straight”. Through the regular laws of sound change, the Modern English words firth and ford are related, as are fjord < Old Norse fjörðr, and port “a haven” < Latin portus. I think Latin vadum “shoal, shallow, ford, sea, etc.” may be related to this same root as well.

The general sense of the Greek word is of a “passage, way, journey”, and it is also connected to the English fare (as in wayfarer and farewell) as well as ferry. It traces its ultimate origins to an Indo-European root √PER meaning “to lead, pass over, pass through” (also the source of prepositions and prefixes of directional meaning: e.g., for(e)– and peri–). This root has all sorts of interesting descendants; not only those previously mentioned, but also such an odd bunch as führer, porter, pier, parsely, fern, feather, gaberdine, and even the proper names Ferdinand, Portugal, and Parvati.

Plato wrote of Poros, a god of expediency, contrivance, and ease (i.e., passage). His antithesis was Aporia, goddess of difficulty, powerlessness, lack of means (i.e., impasse < α + πορία “without passage, means, device”). Aesop and Plutarch each have something to tell us about her. Aporia is a term still used in philosophy to express a state of puzzlement or doubt.

Finally, and I think most significantly, there is the Bosphorus, the Turkish strait that forms part of the boundary between Europe and Asia. The original meaning of the name is literally an ox-ford (βοῦς “ox” + πόρος “passage, ford”). This is amusing to me, and might be to you too, because it recalls the humble origins of the English Oxford and the Four Wise Clerks of Oxenford, the original editors of the Oxford English Dictionary whom Tolkien affectionately parodies in Farmer Giles of Ham.

I also chose the word “boundary” with good reason. If you were paying attention, you noticed I used the same word in the first paragraph of this post. Among drafts and notes for The Lord of the Rings (see The Treason of Isengard, p. 312), Tolkien explicitly identified the River Poros as a “boundary”. In Middle-earth, this was the boundary between Gondor and Harad, but if one overlays Middle-earth very roughly onto a map of our own real world, this corresponds pretty well to the boundary between Europe and Asia, making the Poros roughly analogous to the Bosphorus. Given this analogy and the similarity of the names, the likelihood the Greek word was in Tolkien’s mind seems hard to ignore.

By way of a closing fillip, I’d like to note that this isn’t the first time I’ve speculated about the specific influence of Greek on Tolkien’s nomenclature. If you’re interested and haven’t seen it yet, you might want to read my post on the name of the wolf, Carcharoth. I also wrote the entry on Greek Gods (among others) in the Tolkien Encyclopedia. In the same part of the ancient world, the Colossus of Rhodes seems like a possible model for the Argonath. And I could go on. A culture as rich as that of Ancient Greece could hardly fail to leave traces in Tolkien’s fictive world, especially when you consider that he began his academic career by specializing in Greek philology. A word like πόρος could easily have swum to the front of Tolkien’s mind when he needed a name for a boundary river. This, in fact, could explain why there is no adequate Eldarin gloss for the name. It was all Greek to him. :)


  1. Pssst... Check your title.

    Wouldn't the geographical/geopolitical significance of the Bosphorus depend rather on the real-world era in question? Tolkien might have been thinking of his own present day; or he might have been thinking of eras represented in his philological studies, or those socially and technologically closer to his Legendarium; and the era is going to affect, for example, whether Constantinople is part of Christendom.

  2. It's an intriguing idea. I notice that the drafts of the appendices in HoME XII show that the river's name was tentatively to be Harnen, which I guess means "South-river". Too boring? So JRRT changed it to Poros, with no explanation given by CT as editor. The floor is thus open.

    Yet I'm suspicious of straight transpositions from real European languages to the map of Middle-earth, at least at so late a date as the composition of the history of Gondor. Tolkien repeatedly warned us against such theorizing, as you know.

    So what does Poros mean? Especially because the second part -ros is well-attested Sindarin for a "wet" concept - foam, spray, rain - and is used as such in the nearby Cair Andros, I think I will hold out for Po-, Pho-, Bo-, or Por- being a meaningful element in Sindarin that we just never heard about. CT's comment at the end of "The Problem of Ros" (HoME XII, p. 371) seems particularly strong in support of this idea.

    Thanks for the amazingly erudite expedition, all the same!

  3. My gut tells me you’re right that Poros = πόρος, but not that the River Poros = Βόσπορος.

    I have remarked elsewhere (in a much less scholarly way than is your wont) upon the rather strong and systematic resemblance of Gondor to Byzantium; and within the limits of that analogy, it’s quite plain that the Anduin stands in for the Bosphorus and Minas Tirith for Constantinople. On this analogy the Poros does not represent the border between Asia and Europe, but between the hellenized and unhellenized lands in the East. Ithilien would then correspond roughly to Bithynia and the Thracesians, and Gondor at the time of the War of the Ring to the Byzantine Empire about the time of the First Crusade, when all but a few scraps of its Asian territory had been overrun by the Turks.

    Alexius I sent to the West for mercenaries to help him recover his lost provinces, and was mightily displeased to get a rabble of Crusaders too big for him to police or control; but not as displeased, I imagine, as Denethor when he sent to the North and got nothing but a rag-tag of Hobbits plus token representatives of the other ‘speaking-peoples’. In the end, however, both succeeded at the immediate task; and when Aragorn named Faramir Prince of Ithilien, it might bear analogy to Alexius’ later establishment of the Theme of Neocastra in lands recovered from the Turks.

    By the way, Byzantine studies seem to have been one of Tolkien’s less remarked-upon hobbies; and an interest in Byzantine Greek may have been part of what led him into his acquaintance with the professor of that discipline at Oxford. The said professor, whose name I cannot readily find, is mentioned in Letters no. 17 as having bought a copy of The Hobbit ‘because first editions of “Alice̦” are now very valuable’. Sound commercial judgement, that; but I doubt whether that explanation would have come to Tolkien’s ears if he had not been on more or less friendly professional terms with the man.

    In any case, Tolkien came of a generation in which any boy out of the grammar schools could be presumed to have a reasonable familiarity with Greek — though that presumption was falling out of date by the time The Lord of the Rings came out. He may have stuck Poros in as a sufficiently self-explanatory joke, as he was wont to do: most especially in The Hobbit, with names like Smaug (‘a low philological jest’, Letters no. 25), but to a lesser extent in later works. Sometimes he explained the joke, as with Orthanc, which he explicitly tells us means ‘Cunning Mind’ in the speech of Rohan, but ‘Mount Fang’ in Elvish; but I rather think he did so chiefly because he knew he had to supply the meaning of the Elvish, and not because he supposed it needful to gloss the Old English.

    I doubt very much whether the word Poros can be analysed as containing √ROS-. Gondorian place-names at that period were nearly all Sindarin in form, except for those few that are glossed as ‘pre-Númenorëan’. The Sindarin reflex of √ROS- was roth; it was adopted in that form into Adûnaic, as seen in Rothinzil ‘Foamflower’, the native Númenorëan calque of Vingilot. Since the river was not called Poroth (or Phoroth or the like), I doubt there is any intentional connexion with that etymon — Cair Andros notwithstanding.

    On purely linguistic grounds Cair Andros should have been *Cair Androth. Tolkien was known to modify his names to smooth the reader’s path: thus he chose the irregular form Arnor to avoid the form Ardor, and I suspect he may have given *Androth a ‘Quenyatic’ form to avoid confusion with Andrath.

  4. Trust me to fudge things retrieved solely from memory: of course it was not Alexius I who established the Neocastra, but his grandson Manuel.

  5. Sorry, it’s me again.

    Upon performance of Google-fu, I find that the professor of Byzantine Greek at Oxford in 1937 was one Richard MacGillivray Dawkins, who also did an early study of the Cappadocian language. He and Tolkien must have had much to talk about.

    By the way, Dawkins was the first occupant of the Bywater and Sotheby Chair in Byzantine and Modern Greek Literature. Bywater!! Coincidence? Methinks not!

  6. In Sindarin or in the Mannish language of the area, perhaps.

  7. Interesting post as usual, & I also enjoyed the further thoughts.

  8. I have not much to say about _Poros_ as a name. We have indeed many other names and words from the Elvish languages in the "Lord of the Rings" without a proper explanation (internal Elvish etymology coming from T’s published writings), to name a few: "arth" in Arthedain, Amlaith, Mallor, Rúmil (of Lórien), Orophin, Tarondor.

    I would like to correct Tom Simon. The reflex of the Eldarin root/base ros- is indeed ros, not roth. Final -s stays -s in Sindarin (and in Noldorin-Welsh, too), it does not become th. I can’t remember any instances where Tolkien stated that Andros should be Androth. Noldorin-Welsh/Sindarin -ros and -roth would be two distinct words.

    The explanation provided by Tolkien in his letter nº 347 (Arnor is a mix in popular use of the two names one S. and one Q. ; S. Ardor + Q. Arnanor) is not the same as the external explanation provided in “Words, Phrases and Passages” Parma Eldalamberon nº 17 (p. 28): S. Arnor < reduced from older S. Arannor. I'm still not sure to understand why “Ardor” had to be avoided, according to nº 347.

    Who said that Tolkien "is known to modify his names to smooth the reader’s path"? I would have rather said the contrary, indeed. Unless you're referring to Caras Galadon for "pure/good" Elvish _Caras Galadhon_. According to T, dh looks uncouth to the English eye. (Is it? To the casual modern English speaking reader?). But that makes only one occurence out of a few hundred Elvish names in "The Lord of the Rings" which were not smoothed at all in any special way for the English speaking reader.

  9. If Poros were to be associated with the IE *per- family to which Gk πόρος belongs, it's interesting to note that Tolkien in the QL gives a root PERE with derivatives for "go through", "pierce" ext. to "endure" which I would say was originally meant to somehow reflect that family. This would be supported not only by the coincidence of form and meaning in the root but also by some derivatives like Q perma "passage, aperture" (QL) ~ L porta, portus, Gk πόρος, or Q pere- "endure, experience" ~ Gk πειράω (~ L ex-perior, perītus), etc. Note also Gk πείρω 'pierce'.
    As far as I know other derivatives of this root don't appear in later sources; in the Etym period PER had come to mean "divide in half" (whence perian "halfling") which may be a specialization of the meaning or just a wholly different root, since the semantic group it yields doesn't resemble that of the older PERE root at all. I couldn't say if this means that PERE "go through" as such was abandoned. In any case, in both the QL and Etym (as well as in later sources) a different root TEŘE, TER(ES) "pierce" provides words for "pierce" and "fine, acute", which can easily be related to IE *ter(h1)- "rub" whence Gk τερέω "pierce", τέρετρον "borer, gimlet" (~ L tero, terebra) ~ Q teře- "pierce", teret "auger, borer, gimlet".
    But even if PERE "go through" was not abandoned, Eldarin roots don't show the IE type of vocalic alternation that accounts for the o in πόρος; i.e., it couldn't be explained simply as a Eldarin derivative of this attested root. The relationship, if any, would be more indirect, like the Mannish origin that John suggests.
    On a side note, P. Chantraine s.v. Βόσπορος observes that since it is unlikely that oxen were ever able to cross the strait, the theory that the name owes its origin to the legend of Io (going from Asia to Europe in the shape of a cow) is plausible.
    And since you mention Plato's Poros, it should be remembered that in the Symposium he is united to Penia "poverty, need". This and the verb πένομαι "toil, work; to be poor or needy" can be compared to *PEN "lack, be without" in Q&E and PE17 among other sources. Cross-etymologies crop up everywhere!

  10. R.M. Dawkins "and Tolkien must have had much to talk about."

    To be sure. Dawkins knew Tolkien well; he was one of the Kolbitar, the Icelandic-reading club of dons that Tolkien founded, and he bought a first printing of The Hobbit (for investment purposes). "He liked to discover rare words and enjoyed all variations of dialect," says Maurice Bowra.

    My info is not from Google, but from Hammond and Scull, slightly more helpful sources for the more delicate extensions of information regarding Tolkien.

  11. Tolkien and Dawkins would have found another common ground in the study of folk-stories. Dawkins was a leading scholar in the field of medieval and modern Greek folktales and published extensively on this subject, both collections, translations and critical studies. In his essay "The Meaning of Folktales" (Folklore 62.4, 1951) you can spot some affinities (as well as differences) with Tolkien's "On Fairy-stories" and "Beowulf" essays. Two examples:

    "A second point I would make is that folktales when they have been gathered in a field in which the traditional art was a living affair, are hardly ever stories for children. At most we may say that when the art is dying and people are learning to read books, probably the last stories to die are those that are fitted for children: Puss in Boots and Little Redridinghood are likely to survive such more important and significant stories as Cupid and Psyche and a whole host of moral and moralising tales. Yet these grown-up stories which yield more readily to the advance of literary culture are not less folktales than the children's stories which survive; surviving, partly because children don't read and still more because they like hearing stories rather than reading them." (417-8; cfr. the "Children" section of OFS.)

    "It was a common practice in medieval Europe to use allegorical interpretations to give a spiritual, a wider and a deeper meaning to almost any story, even the most apparently refractory; to add what the writer of the Gesta Romanorum calls a moralitas. But the symbolic element which I would see in so many folktales is hardly the same thing, though it may serve the same end, that of providing a story with a deeper interest. The medieval moralitas is a later addition to the story added by way of edifying comment or explanation. In the case of a myth, conveyed in a story or even in a parable, the hidden meaning is in the story from its very conception; it is not in the least a case of adding one thing to another. The concrete, external form of a story is used to convey ideas and feelings: in a parable by taking thought; in a myth by a far less intellectual process. The poet could no more give an account of the relation between the hidden value and the external form of his story than the hearer could say in terms why such a story had given him satisfaction." (428; cfr. "The significance of a myth is not easily to be pinned on paper by analytical reasoning. It is at its best when it is presented by a poet who feels rather than makes explicit what his theme portends" etc., MC:15)

  12. Thanks to all of you for the valuable comments! A few quick replies (though you are all doing very well on your own!):

    @Robert: I corrected the typo in the title — thanks very much for alerting me to that. D’oh!!

    @squire: Yet I’m suspicious of straight transpositions from real European languages to the map of Middle-earth, at least at so late a date as the composition of the history of Gondor. Tolkien repeatedly warned us against such theorizing, as you know.

    Indeed, and I am pretty careful about offering up such hypotheses. It is still purely speculative, but I think there’s enough (circumstantial) evidence to be worth sharing the idea. By the way, though Tolkien repeatedly warned against it, he did this kind of thing himself, also repeatedly. Example: “I went for a brief holiday in Gondor (or in modern terms Venice)”. Sure, he’s kidding around, but this certainly implies an overlapping of real and fictive maps in his mind (though not necessarily of languages).

    @Hlaford: If Poros were to be associated with the IE *per- family to which Gk πόρος belongs, it’s interesting to note that Tolkien in the QL gives a root PERE with derivatives for “go through”, “pierce” ext. to “endure” which I would say was originally meant to somehow reflect that family.

    Nice to see you again, Hlaford. Thanks very much for pointing this out (as well as everything you wrote after). I did check the Gnomish and Qenya Lexicons, but it was obviously too hasty a peek, and I failed to notice this root!

  13. @E. J. Kloczko:

    Of course you are right, and I have the contents of a box without hinges, key, or lid all over my face. Most of my reference library is packed after a recent (and chaotic) move, and I foolishly spared myself the effort of unpacking enough to check the mutations of word-final S in Sindarin.

    Also I was led astray by the Adûnaic, which is clearly shot through with loan-words from Sindarin: roth is unquestionably one, and so, for that matter, is adûn. (Also, I strongly suspect, ar and [a]khor, from ar[an] and hîr respectively. It was a fine stroke of irony to make Ar-Adûnakhor outlaw the use of the Elvish tongues, and then take a regnal name patched together from three notorious Sindarin loan-words!)

    However, in re Ardor: The reason Tolkien wished to avoid it is that it is a common English word, and would look ridiculous as the name of a lost kingdom; it would seem like pure allegory, something belonging in Bunyan and utterly inappropriate to Middle-earth. What he would have made of the (later) English meaning of, say, Teleporno, I shudder to think.

    @David Bratman:

    I don’t happen to have Hammond & Scull handy, or indeed in my personal collection, such as it is. I am not a professional Tolkienist and can seldom afford to buy books not directly needful for my own work. In fact I was going solely from the mention of ‘the Professor of Byzantine Greek’, not from a members’ list of the Kólbitar. Really, you make me feel like Gandalf when Thorin chewed him out for taking so long to bring him Thrór’s Map: ‘Your father did not know his own name, and he never told me yours.’

    I apologize to you all for wasting your time.

  14. I love these word games very much - not least because they are both entertaining and enlightening.

    Since my training in Greek is limited to the use of Greek letters in physics and mathematics and my knowledge of Sindarin is only marginally better, I will refrain from offering any new comments on these :-)

    Speaking more generally, I often find that there is a tendency to attribute a level deliberation to Tolkien's work that seems to me unrealistic: his comment that hardly a word in The Lord of the Rings has been unconsidered does not, I think, mean that names had necessarily been considered to the point of Tolkien having thought through such elaborate word-games. There has been a tendency by some of the leading Tolkien scholars to warn us against trusting Tolkien's own statements too literally, and I think that this also applies to this statement.

    The evidence shows, I believe, that when Tolkien needed a new name in his narrative, he would often merely invent a name on the spot that merely sounded right, and though he would also often come back and invent an appropriate internal etymology for the name, that wasn't always the case.

    Jason has already mentioned ‘The Rivers and Beacon-hills of Gondor’ and the hill Erelas seems an example of this. Tolkien notes that it is ‘Sindarin in style, but has no suitable meaning in that language.’ (VT42 p.19 — emphasis added). The impression here is that Tolkien invented a new Sindarin-sounding name on the spot, only to later discover that the implied meaning (Single-leaf?) was unsuitable.

    Casting for such names, it would be surprising if he did not at times settle on words from Primary World languages that he knew, and it is unlikely that this was in all cases deliberate (prof. Olsen's discussion in one of his podcasts of Túna is relevant here) — I actually doubt that it was in most cases initially deliberate or conscious, though he of course would often discover it afterwards.

    Poros could, for all I know, be a deliberate loan from Greek, an unintentional and possibly even consciously recognized loan, a mere coincidence or a deliberate construction in Sindarin, Quenya or some other invented language. I think all of these possibilities can be found in Tolkien's work, and I know of no evidence that could finally show which is correct.

  15. This is great stuff, I could read this sort of thing all day, but sadly under the auspices of the Goddess Aporia since my own linguistic knowledge is sparse. Thanks guys

    (From Saranna - the blog won't let me in again though it did five minutes ago! *sigh*)

  16. Very well said, Troels! I’m glad you enjoyed the post as it was really intended — in a spirit of light-handed (if not light-minded) wordplay.

    Of course, you are absolutely right: we have no way of knowing whether it was a borrowing from Greek or not, and even if so, whether it was intentional or not. Not unless further information should come to light. For me, though entirely circumstantial, the evidence is still pretty convincing. More so than, say, my ruminations on Khamûl or Radagast. :)