Middle-earth seen by the barbarians: The complete collection including a previously unpublished essay

by Codex Regius

  Dorwinion remained the east-march of Gondor for more than a millenium, until 1856 TA, (TY) when the Wainrider empire, stirred by the power of Dol Guldur, expanded to cut it effectively off from its mainland (see chapter III.3.3). ‘The people of eastern and southern Rhovanion were enslaved; and the frontiers of Gondor were for that time withdrawn to the Anduin and the Emyn Muil.’ (KR) Not only ‘for that time’: They never returned to the Sea of Rhún.

  Lost by Osgiliath, the exclave of Dorwinion will have decided to autonomously sustain itself. The then vacant throne may have been claimed by local nobility, still of mingled Dúnedain descent, who continued the traditions of their old lords like choosing ascension names from Elvish languages (unless this was a specific honour to their trading partners, ‘the Sindarin Princes of the Silvan Elves’, see GC). This tradition continued unbroken till the reign of Bladorthin in 2770 or even after, and Dorwinion was never subdued nor was its aristocracy routed or expelled by the Wainriders, the Balchoth or other subsequent Easterling invasions.

  The survival of the little Dúnedainic exclave that vanished, unheeded, from Gondorian records may have been supported by the particularities of its territory. The military forces of Rhún, like the Wainriders, relied on chariots (as their name implies) or, more rarely, on cavalry units. This strategy is suited for level battlefields where such forces can take advantage of their mobility. But on sloped or elevated territory, both chariots and riders are in a disadvantage against resolute defending partisans. The hill range of southern Dorwinion was almost inaccessible to troops of this kind. There the defenders were able to hold out by guerilla warfare for a long time. No wonder then that the people of Dorwinion survived the turmoil of the late Third Age unscathed, enjoying long periods of peace and safe commerce. Even after they had gained sovereignty - welcome or not -, their land stayed so much westernised that, the only one among the states and realms between the Withered Heath and the Ash Mountains, it was still known by its Sindarin name.

  Bladorthin, therefore, was most likely one of the most recent kings of Dorwinion, a man of mixed Dúnedain and Northmen descent. What did he do to earn particular honours as ‘the great king’?

  Since he was attestedly alive in 2770, it is plausible to assume that he was on the throne already in 2758, that fatal year when Gondor and Rohan were under simultaneous attack by Easterlings and Southrons, when the Long Winter held Middle-earth in its grasp and when there was certainly need for the skills of a great king East of Anduin. Bladorthin may have gained his merits by defending Dorwinion against the Easterlings and by coping with the effects of the Long Winter: a feat that was certainly worth to be recalled in ballads and memories. When this had been achieved, around the year 2769, Bladorthin saw the need to replenish his armories with the skills of Dwarvish smiths so that he could further defend Dorwinion’s sovereignty against the persistent menace from Rhûn - a project that, owing to unforeseen circumstances, would remain unfinished.

  If all that is true, then one question will remain unanswered. When in the Fourth Age king Elessar reclaimed Dorwinion as the east-march of Gondor, was he welcome there?

  Of course it cannot be known whether J.R.R. Tolkien, had he ever thought about Bladorthin again after his brief mentioning in H, would have made him king of Dorwinion. But this is definitely the most plausible space where to integrate him into the history of the Third Age.

  Appendix A: The meaning of the name Bladorthin

  It is a common assumption that the name of king Bladorthin was related to that of Bladorwen, a rejected ‘Gnomish’ name of the Vala Yavanna in the very early wordlist of GL. Since Bladorwen is translated there as ‘the wide earth’, some people have simply added Gnomish and Sindarin, blador + thin (as in Thingol, ‘Grey-cloak’, HH1 chooses to point out), and get ‘Earth-grey’ as a result. That does not make much sense as the throne-name of a king, does it?

  Considering that in the early drafts of H, Bladorthin was not yet an ominous king but the very wizard who would later become Gandalf, ‘the name could even be interpreted as “Grey Wanderer” …, thus becoming an early precursor of Gandalf’s … elven name, Mithrandir.’ (HH1) Which is, of course, actually translated as ‘the Grey Pilgrim’.

  But this assumption fails on several grounds. First, the main Elvish languages in use when H was written (that is, in 1937/38) were Q(u)enya and Noldorin, the latter had developed from the early vocabulary of Goldogrin/Gnomish established in GL while early Qenya is found in QL and Early Noldorin in NW. It is there that we should look for the proper meaning of the great king’s name.

  BLAD

  GL contains several very suggestive entries. They are rearranged here for purposes of clarity:

  blath a floor.

  blant flat, open. expansive. candid, blunt (of words, etc.).

  bladwen a plain.

  Bladorwen the wide earth. The world and all its plants and fruit. Mother Earth.

  GL includes a few further derivations that discussions of the word-group PAL usually ignore:

  ‘blantos (= mavlantos) sycamore.’ - ‘mavlantos sycamore’.

  ‘mavlant = mablod.’ - ‘mablod palm (of hand).’

  These words are obviously compounds of mô ‘hand’ and blant ‘flat, open’. They are mostly ignored because readers do not understand the implications, it seems. The sycamore referred to in the above is not what Americans think it is. Europeans mean acer pseudoplatanus by that, a common alpine tree whose leaves feature five vaguely finger-like protrusions. That observation is why the Gnomes dubbed it *flat-hand tree or, more literally. *palm-tree!

  There are certain evident regularities. The basic structure is this: a base *blada- that produces an adjective blant and the nouns blath, blador, bladwen. HH1 goes wrong in the assumption that the final -wen was a female suffix. Gnomish has various complements to the blath/bladwen pair that do not imply gender:

  faith/faidwen, ‘liberty/freedom’, respectively;

  flad/fladwen (and fladweth) ‘sward, grass’ vs. ‘meadow, grassland’;

  gados(< gador)/gadorwen ‘union, association, fellowship’ vs- ‘society’ (and gadweth ‘union, joining’) from a base *gada- ‘join, connect, unite’ etc.

  In general, Gnomish -wen seems to imply a larger or even all-encompassing dimension than the plain word without this suffix.

  QL adds a few corresponding words under the entry PALA:

  palanta even, flat, level.

  palo(u) plane surface, plain, the flat.

  Palurin, Palurien the wide world.

  palapapte = mapalin. (Mapalin is not glossed but seems to be the Q equivalent of mavlant, the sycamore).

  But the most remarkable entries are these:

  palwa- make wander. palava wandering. palava- to stray, wander.

  May HH1 yet be right about the ‘Wanderer’? This elaborate vocabulary is supplemented by a single entry made to the next stage of linguistic development, Early Noldorin:

  blador world (NW).

  No doubt that this element was also present in Gnomish, though it is not listed in GL. The difference to Bladorwen is probably gradual: The suffix -wen suggests the entire cosmos, ‘the wide earth’, while the basic blador is, apparently, just ‘*the world around us’

  THIN

  The suffix -thin is believed by most interpreters to be a cognate of Sindarin thin “grey”. John Rateliff, in HH1, bases this assumption on mere speculation: ‘I have not found a gloss of “Thingol” from this early period, but there is no reason to doubt that it would have been the same as the later Sindarin translation: “grey-cloak”, with thin = grey.’

  Well, the gloss that Mr Rateliff was missing is in fact present in NW, and it does not read ‘grey’:

  ‘thing, prince; thingol [=] Singoldo.’

  It seems pretty obvious to me that Bladorthin is a contraction of Blador + thing, i. e. ‘*World-prince’, ‘Prince of the World’. This is a boasting name for a great king but certainly not one unheard of. At least it is much better than that assumption
by a German critic who asserted that changing the wizard’s name from Bladorthin to Gandalf had been a first step of advancing H to the more serious mood of LR. After all, Bladder-thin was not really a suitable name for an Istar, was it?

  A letter published in L contains this remarkable passage:

  ‘If Hobbiton and Rivendell are taken (as intended) to be at about the latitude of Oxford, then Minas Tirith, 600 miles south, is at about the latitude of Florence. The Mouths of Anduin and the ancient city of Pelargir are at about the latitude of ancient Troy.’ (L294)

  To be honest, this statement sounds a bit unlikely. A modern atlas will quickly show that it was extremely vague, to say the least. The latitudes of Florence and Troy are twice as much apart as those of Minas Tirith and Pelargir (verifying the scale of the large LR foldout map against the one reproduced in TR). However, once we accept the latitude of Hobbiton, a meridional grid may be applied to the Middle-Earth map so that it may be projected on a globe (a feat that Karen Fonstad thought impossible, cf. TAME). The results of this projection are shown in Fig. 25 et seq.

  The following discussion is based on the original LR maps, printed before the publication of UT.

  Appendix B: Latitudes and longitudes of Middle-earth

  To apply latitudes and longitudes, we have to know what kind of cartographic projection the LR map is based on. Unfortunately, the published map shows both a scale and a compass rose. And you cannot have both of them on a map unless you assume that you are depicting Arda Flat of the Second Age. Projecting a round earth always demands that either your scale or your directions are distorted!

  Fortunately, the addition of the compass rose turns out to be a fault made only in the published version. Tolkien’s original hand-drawn maps in RS and TR show a scale but not a compass rose. Instead, a rectangular grid is superimposed on both maps. Each square has a vertical and horizontal side length of 100 miles (hence, on the originally published LR map, each square measures exactly 2 cm, which is certainly not an accident).

  This is a cartographic projection that permits a geometric transformation into latitudes (parallels) and longitudes (meridians). A few additional assumptions are required:

  One meridian should run due North-South, preferably in the middle of the map. Incidentally, the central line of the published LR map passes straight through Rivendell. That is perhaps no accident, either.

  All horizontal grid lines on the HOME maps are running due East-West. On a globe, they would be parallels to the equator, mapped at the same scale as the central meridian. For at least one grid line, this assumption is confirmed by the statement quoted in the beginning: Both Hobbiton and Rivendell are found at the same latitude as well as on the same horizontal grid line.

  The distance of the horizontal lines is constant. This is confirmed by RS: the distance from one cross of the grid to another, either horizontally or vertically, is said to be 100 miles.

  This evidence allows to deduce the projection of the HOME maps. The latitudes are parallel to the horizontal grid lines while the longitudes converge towards the upper edge of the map sheet (towards the north pole). This means that the published LR map preserves distances but not areas and directions and it is the compass rose that should be deleted, not the scale: the angles are increasingly distorted towards the upper left and right corners of the sheet. Only the central meridian is in fact running due South-North; the other longitudes describe paraboloids that, hypothetically, converge with the central meridian at the poles.

  The scale of the published LR map is 1 cm = 50 miles or 1:8 000 000. On the earth, two degrees of latitude are about 111 km apart from each other. Two adjacent meridians are separated equally far at the equator, to converge towards the poles. Hence, for a known latitude L, the distance D from one crosspoint of a meridian to the next crosspoint (in steps of one degree) can be computed as:

  D = 111 km * cos L (L in deg., D in km)

  If you want to know how many centimetres that is on the LR map, remember that 2 cm = 100 miles, so that 1.25 cm = 100 km. Thus, the distance D1 of two crosspoints, measured in cm, is:

  D1 = D * 1.25 cm/100 km

  D1 = approx. 1.4 cm * cos L (L in deg, D1 in km)

  For a given latitude, start at the central meridian of the LR map and count the crosspoints with the proper longitudes. This will be done in the following

  Latitudes on the LR Map

  Oxford at located at about 51°45’N. Accepting this value as the latitude of Hobbiton in accordance with L294, we may now draw the proper parallels on the LR map. The main parallels on the LR map and a modern world map follow.

  35°N - The Western exit of the firth of Umbar

  In Europe/North Africa: Sahara-Atlas, Crete, Cyprus

  The City of the Corsairs is found at about the latitude of Tanger.

  40°N - Southern Tolfalas, the W-E range of the Ephel Duath

  In Europe/Asia: Madrid (Spain), Sardinia, Gulf of Tarent, Troy

  Contrary to Tolkien’s estimation, Pelargir is located at 41°10’, at the latitude of Istanbul rather than of Troy.

  45°N - Tol Brandir, Dead Marshes, Morannon

  In Europe: Bordeaux (France), the plain of the river Po, Crimea

  Minas Tirith is located almost exactly at 43°. But this is not the latitude of Florence, 43°50’, that would be closer to Caer Andros.

  50°N - Eregion, the East Bight, the lower W-E flow of the Carnen

  In Europe: Lizard Point/Cornwall, Amiens, Mainz, Bohemia

  In Beleriand, this parallel would continue via Ossiriand between Duilwen and Adurant to the sand bank in the Bay of Balar. But in the First Age, earth was flat and a different coordinate grid applied.

  55°N - Mount Gundabad, Ered Mithrin, Withered Heath

  In Europe: Londonderry, Fuenen, Bornholm, Lithuania

  On the RS map, this latitude corresponds to the uninterpreted line that bisects the squares H and I.

  In Beleriand, the line would pass Lake Mithrim, Barad Eithel, Haudh-en-Ndengin.

  Longitudes on the LR Map

  The meridians follows. As mentioned before, the central meridian seems to cross Rivendell where the LR map is supposed to have originated. The ‘Greenwich’ meridian of Middle-earth could arbitrarily be applied there.

  But there may be a different one that separates West and East: ‘Its eastern frontier was roughly the River Carnen to its junction with Celduin (the River Running), and so to Núrnen, and thence south to the ancient confines of South Gondor. (It did not originally exclude Mordor …)’ (TI) In this system, the Rivendell meridian is 13°W, resulting in the following:

  35°W - The island of Himling.

  30°W - Northern Ered Luin, Harlindon, missing Harlond to the west.

  25°W - Forochel, via Hills of Evendim and the Far Downs, to the Mouths of Baranduin.

  20°W - North Downs, Midgewater Marshes, Enedwaith, Andrast.

  15°W - Ettenmoors, Eregion, Orthanc (precisely!), Anfalas.

  10°W - Missing the Carrock a little to the east, the Field of Celebrant, the tip of the Entwash delta, Harondor.

  5°W - East of the Mountains of Mirkwood, eastern edge of Udun.

  0° - Carnen, Celduin, Nurn.

  5°E - The shallows in the Sea of Rhun, passing the confines of Mordor in the East.

  Projected on the map of Europe, the LR map covers Central Europe and the Mediterranean as expected. Its area spans 20° of latitude (55 to 35°) and, at the northern edge of the map, up to 40 ° of longitude (29°W to 10°E).

  If we assume that Hobbiton is not only at the latitude but also at the longitude of Oxford, then the LR map overlays the geographic territory between Scotland and Crete and between Ireland and Kiev. Minas Tirith and Osgiliath are now submerged in the Adriatic Sea at 43°N, 17°E, somewhat off the coast of Split (Croatia). Barad-dûr is located near Belgrade - some have said, right under the desk of the Serbian president.

  Appendix C: Pauline Baynes’ annotated map

  In October 2015, an LR map full of hand-writ
ten annotations by J.R.R. Tolkien was found in a copy of the books owned by Pauline Baynes. It was evidently the working draught that inspired Baynes’ full-colour map of Middle-earth, published in 1969. This map, PBM, notoriously included a number of names that were not found on the map included in LR and it had been supposed that these additional terms had been communicated to Baynes by Tolkien. The draught, PBD, fully confirms this previous assumption.

  Tolkien’s often hard to read notes on PBD, published in 2015 by Blackwell, have since been transcribed by Blackwell’s employee Susan Theobald to reveal various intriguing new details which affected the discussions in this book.

  First of all, in the lower left corner there is a statement found to the extent that ‘Hobbiton is assumed to be approx. at latitude of Oxford’. This is not as new as many commentators have thought. Tolkien simply repeats his statement made in L294 that has been quoted in Appendix B. Unfortunately, he failed to add, again, whether Hobbiton was also supposed to be located at the longitude of Oxford, i. e. whether or not it is a geographical, if not historical, precursor of modern Oxford.