Some email-discussions already emerged concerning the essay Mapping Arda before the online-publication of OtherMinds 01. I will now use this public forum to summarize them for the broader public.

Why did we consider Tolkien's Ambarkanta-map-V more than Ambarkanta-map-IV?

Tolkien very carefully drew Ambarkanta-map-IV in ink in the year 1936. Perhaps a few months later he drew another map at the back of it, Ambarkanta-map-V. But the later cannot compete with the former, because it was very roughly and hastily sketched in pencil.

If you compare our maps with Tolkien's Ambarkanta-maps, you might recognize at once that our maps are based more on Ambarkanta-map-V than on Ambarkanta-map-IV, although map-V is not more than a rough sketch. But we had reasons to do so and I will now try to explain them.

a) Timeframe of the map

Our First-Age-map is mainly supposed to be used for gaming and story-telling during the First Age of the Sun. Thus it shows the geography after the three long Ages of the Stars. During those Ages the Battle of the Powers had taken place, Utumno was destroyed and Melkor was taken captive.

On Ambarkanta-map-IV, Tolkien shows the geographic situation after the fall of the Lamps as he imagined in 1936. The map shows the landscapes still long before the Battle of the Powers. On the other hand, Ambarkanta-map-V was explicitly drawn to show the geographic situation which came into being after Utumno was destructed.

Since Ambarkanta-map-V visualizes the geographic situation between Ambarkanta-map-IV and our own First-Age-map, we were simply not able to ignore it even if it looks sketchy.

b) Working with sketchy sources

As far as Tolkien is concerned, big chunks of the primary sources simply are in a very sketchy fashion. Ever since they were painstakingly deciphered and published by Christopher Tolkien in Unfinished Tales and History of Middle-earth, we got accustomed to work with sketches, to weigh them against each other and against older (but more legible) manuscript(layer)s and typescripts. This mental process has been named canonization. It already began with Christopher Tolkien himself as he chose between the various sources to edit his version of The Silmarillion in 1975:

"On [amanuensis typescript] LQ2 [1958] my father emended Vaiya to Ekkaia (whence its occurrence in the published Silmarillion)." (HoME 10: The Later Silmarillion, comment on §12)

First of all, our maps are a work of canonization. As a rule we canonized the available primary sources the following way: We searched for the last hint Tolkien gave about a certain geographic topic. Then we transformed it into a map-feature. If you read through our Essay, you will find many instances of canonization.

Since Ambarkanta-map-IV is older than Ambarkanta-map-V, regularly the later seemed more valid to us. There's one exception, though. In 1946 Tolkien wrote various versions of The Drowning of Anadûne (HoME 09). In those writings we found hints that in 1946 the eastern continent of the Empty Lands should again look more like on Ambarkanta-map-IV than on Ambarkanta-map-V. Hence we had to conclude that on our own maps the shape of the Empty Lands should be in line with the older Ambarkanta-map-IV.

The extent of eastern Middle-earth

As Tolkien repeatedly said, his stories were imagined to take place in a kind of mythological past of the isle of Britain, respectively western Europe. Hence Tolkien had to develop a mode of transition between the fantasy-geography of Beleriand and the modern geography of western Europe. This is exactly what is shown on Ambarkanta-map-V. There you can see Beleriand together with the continent of Africa and a somehow halved version of Asia.

Naturally this illusion-of-transition seemed no longer possible after Tolkien merged the geography of Beleriand and Wilderland in 1938. Big geologic structures like the Misty Mountains or the Grey Mountains are simply not vanishing even in 200 million years.

Since 1938 there was no serious way to transfer western Middle-earth into western Europe. The geography of both lands had become too dissimilar. Hence Middle-earth should be more regarded as a mythological-but-impossible past than as a magical-possible past of Eurasia/Africa. For us, this seems the only way not to overstress the wilful suspension of disbelief.

Because of this incompatibility Pete Fenlon was free to reinvent the geography of eastern and southern Middle-earth in a like incompatible way. He had not to borrow from the geography of real-world-Asia or Africa. Instead he just took certain features from the Ambarkanta-maps (e.g. Iron Mountains) and mixed them heavily with his own ideas.

But there's a further feature of Ambarkanta-map-V to which Fenlon stayed true. This was that somehow halved size of Asia, sharply cut after the peninsula of India: In 1937 Tolkien might really have imagined that in those mythological past the continent of Asia was just half as big as it is now.

As far as we know, after 1937 Tolkien never again uttered any precise information concerning the size of eastern Middle-earth. His last comments in this question were the following:

"The Great Central Land, Europe and Asia, was first inhabited. Men awoke in Mesopotamia. Their fates as they spread were very various." (HoME 09: Sketch I of the Drowning of Anadûne; 1946)

"Men 'awoke' first in the midst of the Great Middle Earth (Europe and Asia), and Asia was first thinly inhabited, before the Dark Ages of great cold. Even before that time Men had spread westward (and eastward) as far as the shores of the Sea." (HoME 09: Sketch II of the Drowning of Anadûne; 1946)

"In the Dwarvish traditions of the Third Age the names of the places where each of the Seven Ancestors had 'awakened' were remembered; but only two of them were known to Elves and Men of the West: the most westerly, the awakening place of the ancestors of the Firebeards and the Broadbeams; and that of the ancestor of the Longbeards, the eldest in making and awakening. The first had been in the north of the Ered Lindon, the great eastern wall of Beleriand, of which the Blue Mountains of the Second and later ages were the remnant; the second had been Mount Gundabad (in origin a Khuzdul name), which was therefore revered by the Dwarves, and its occupation in the Third Age by the Orks of Sauron was one of the chief reasons for their great hatred of the Orks. The other two places were eastward, at distances as great or greater than that between the Blue Mountains and Gundabad: the arising of the Ironfists and Stiffbeards, and that of the Blacklocks and Stonefoots. Though these four points were far sundered the Dwarves of different kindreds were in communication, and in the early ages often held assemblies of delegates at Mount Gundabad." (HoME 12: Of Dwarves and Men; 1969)

Hence Ambarkanta-map-V shows Tolkien's last explicit thought about the size of eastern Middle-earth during the time of his stories. And in that mythological time it was not nearly half as big as modern Asia.

Consequently Pete Fenlon drew his version of eastern middle-earth in the same shrunken size. And for our own maps we follow him as well as we follow Ambarkanta-map-V.