Can we get three questions out of the way before we start?
1: Ought Tolkien to have engaged in world-building above and beyond what was needed to tell a story?
2: Given that he did, ought his relatives to have made his unpublished notebooks available to the public?
3: Given that they did, ought we to read them?
The answers being: yes, if he wanted to; yes, if they want to; and yes, if we want to.
The Nature of Middle-earth is a sad, infuriating, fascinating and above all, unreadable work. If it were all we had to go on, we would describe Tolkien's creative process as a pathology. We would place him, at best, alongside the more obscure gnostic ramblings of Phillip K. Dick; and at worst, alongside the patterns some crazy man has daubed on the wall of his padded-cell: meaningful to him, meaningless to the rest of us.
But this book is not all we have to go on: at some level, these kinds of writings -- jotted down on the backs of exam papers and royalty statements, yet carefully preserved by Tolkien himself and by his children -- led to the Tale of Beren and Luthien, to the Mirror of Galadriel, to the Steward and the King. Can we make the connection?
Many of us doodle on the backs of envelopes. The Complete Doodles of Leonardo Da Vinci or the Collected Scribbles of Van Gogh would be intensely interesting. They might not be great works of art; but they would tell us things about the artists' development and their working practices that we couldn't find out anywhere else.
The Complete Doodles of Andrew Rilstone, not so much.
The Nature of Middle-earth doesn't tell us much about the nature of Middle-earth. But it does tell us a good deal about the nature of Tolkien's creative process. More, perhaps, than we actually wanted to know.
There is no single, finished thing called Middle-earth to talk about the nature of; only three differently unfinished works in progress.
There is, if you will, Middle-earth I, the setting of the Book of Lost Tales, back when Beren was an Elf, Sauron was a cat and minstrels had names like Tinfang Warble.
There is Middle-earth II, the world of Lord of the Rings and the published Silmarillion, when Hobbits, Dwarves and the sunken island of Numenor had inveigled themselves into the long-standing Elf-mythology.
And there is the projected Middle-earth III which would have made the world of Lord of the Rings more consistent with real-world geography, real-world astronomy and real-world theology. It would have ret-conned out the flat-earth, the sky done, and the literal sun-chariot, and made Eru and Morgoth theologically consistent analogues for the Catholic God and the Catholic Satan.
Maybe Numenor-Atlantis never sunk beneath the waves, muses Tolkien at one point. Maybe it just had all the magic sucked out of it and turned into America.
The notes that make up the Nature of Middle-earth were written at the very end of Tolkien's life; and are therefore mainly about the projected, unfinished, and inferior version of what purists call The Legendarium. Anyone expecting a collection of geek-heavy lore -- more or less canonical information about the setting that got left out of the books -- is going to be deeply disappointed. Very little in this book would be of the slightest use to someone producing, say, a role-playing game, a work of fan-fiction, or a four hundred and sixty five million dollar TV series for Amazon Prime. It would be more help, I think, to someone trying to write a grammar of Elvish; although even here Tolkien's thinking is in a permanent state of flux. Carl Hostetter, the editor, is an expert on Quenya and Sindarin and used to be the head of the Elvish Linguistic Fellowship. He also wrote computer software for NASA, which was probably a lot more straightforward.
In the Silmarillion, the Elves are created and wake up in Middle-earth long before the creation of the sun and the moon; they go on a very long journey to Valinor (where the gods live under the light of the Two Trees) but they eventually rebel against the gods and come back to Middle-earth. Tolkien has decided that for Lord of the Rings to work as a fictional pre-history of our own Earth, the sun and the moon must have always existed. And this puts his chronology right out. He starts doing sums in his head about how long a Valinorian year lasts compared with a "year of the Sun", how long Elves live for, and how much they breed. Given that they are immortal, if he gets his figures wrong the world is going to be teeming with the little pointy-eared buggers; if there are that many, they can't all have decamped to Valinor and he doesn't want to spoil the story about how the first Men found quite a small tribe of Elves when they first came over the mountains.
And so: pages and pages of different "generational schemes": if Elves hit puberty at such-and-such an age and have yay many babies, what would the population be after X, Y or Z generations? At one point he calculates the ratio of Elf-Years to Human-Years to three hundred decimal places -- without, the editor reminds us, the aid of a pocket calculator or correcting fluid.
This is not writing; this is not world-building. It isn't even doodling. It is displacement activity.
Elves live longer than humans, but not ridiculously longer: not much more than a hundred and fifty years. But time passes much more slowly in Valinor, and once they are in Middle-earth, Elves continue to age at the Valinorian rate (although they perceive time as humans do). Arwen is 2788 years old at her marriage, but, subjectively, she's only about 28.
Naturally, it's not that simple: Elvish children age at about the same rate as human children; and hit puberty at about the same time. And age doesn't effect Elves the way it does humans. They don't go grey and wrinkly or become weaker and they don't grow beards. Their spirits (fea) and their bodies (hroa) are plumbed together very differently from those of humans. At their creation, Elves looked very much like men; but the longer they live the more spiritualised their flesh becomes. In the end, their fea "uses up" their hroa: any Elves in Middle-earth today (in the twenty first century) would have bodies that have become completely spiritualised, and would appear to us as disembodied or invisible beings. Making babies exhausts some of their fea, which is why they (fortunately) tend to have very small families.
Hostetter provides a helpful appendix pointing out that this is in line with Catholic thinking. The idea that the soul inhabits the body and is separable from it is a gnostic heresy. A good Catholic thinks that a human person is a unity of body and spirit. (The theology of the incarnation of Jesus became very complicated around this point.) Tolkien's theory that the fall of Morgoth-Satan entailed a corruption in the nature of they physical universe depends on this Thomist/Platonic conception of matter.
So what happens if an Elf is killed? Tolkien can't simply say that an Elf's spirit goes to Elf-heaven: he is very committed to the idea that Elves, body-and-soul, are "coeval" with the life of the material universe. It takes a direct conversation between Eru (literally God) and Manwe (top Valar on earth) to sort it out. Eru gives Manwe permission to create new bodies for any Elves who have been temporarily dis-incarnated. Tolkien gets justifiably worried about whether an Elf with a new body can be said to be the same elf as he was before, and goes down a deep rabbit hole about whether an exact replica of your house would still be your house. (Yes, if your house burned down and was replaced by one exactly the same you would feel that you had your house back; except with respect to an object that you loved particularly because it was, say, a gift from a friend. That can't be replaced because what you love is not the thing itself, but its history. And so on for some pages.)
In places, Tolkien's lyrical story telling voice does shine through. Here is is recounting the Elves own story about their first generation:
Imin, Tatis and Enel awoke before their spouses, and the first thing that they saw was the stars, for they woke in the early twilight before dawn. And the next thing they saw was their destined spouse lying asleep on the green sward beside them. Then they were so enamoured of their beauty that their desire for speech was immediately quickened and they began "to think of words" to speak and sing in. And being impatient they could not wait but woke up their spouses. Thus (say the Eldar) elf-women ever after reached maturity sooner than elf-men; for it had been intended that they should wake later than their spouses.
The six go exploring and find another group of twelve; the twelve find twenty four, and so on until there are a hundred and forty four of them. The Elvish words for the numerals one, two and three (Minn, Atta, and Nel-De) come from the names of those first three Elves, although scholar-Tolkien adds a footnote saying that this is a story-internal myth; and that it is more likely that latert Elves retrospectively named their first ancestors after the numerals. Because there were twelve dozen Elves in the original creation, the Elves count in base twelve and don't have names for any number higher than a hundred and forty four. (A hundred and forty four is, of course, a significant number in the first chapter of Lord of the Rings.) Elvish children play a this-little-piggy game in which they call their fingers Daddy, Mummy, Big Boy, Little Girl and Baby. Those words also sound like the numbers from one to five.
Tolkien has started with some theoretical ideas about the origins of language; some mathematical calculations about how many Elves there need to have been for the demographics to make sense; and even a note about the rate of maturation in he-Elves and she-Elves: but the ideas have come together into a charming narrative about Elf-Adam waking up Elf-Eve too early because they were so keen to start creating poems. That particular story would not have happened if not for the theoretical ground-work.
Some of the material is of more general interest. Tolkien is quite annoyed by Pauline Baynes' illustration for a poster-map of Middle-earth; he takes particular exception to her giving the Nazgul hats: so he provides extensive notes about what the characters looked like. He seems to be interpreting his own text as if it were an historical document. She shouldn't have drawn Gollum going naked, because the text says he had pockets; he "evidently" had black garments, but when the text says he is "as dark as darkness" that just means he couldn't be seen with normal eyes in a dark cave. Elsewhere he explains that Aragorn and Boromir would not have had beards, not because they shave, but because of their Numenorean ancestry and Elvish blood: Elvish men don't grow beards. (This made we wonder about the beardlessness of Hobbits.)
But some of it is really astonishingly trivial. A paragraph about the Druedain's cultivation of mushrooms (rejected by Tolkien because it made them too much like Hobbits) is lovingly reproduced. We learn that they liked fungi; that they knew which were poisonous and which were not; that they planted fungi near their dwellings; that Elves on the whole don't like mushrooms; that some people even think that fungus has been cursed by Morgoth. It's like Tolkien feels that he has to say something about everything, even if he has nothing to say. But as a result of the doodle, he now knows that some people in Middle-earth call mushrooms "orc plants".
Or again: Tolkien makes a linguistic note about the Elvish word gor, which can mean "warn" but also "influence" or "counsel". To elucidate the nuance, he says that it is derived from the word ore, which in Lord of the Rings is translated as "heart" ("my heart tells me") but which does not in fact refer metaphorically to the bodily organ. It means something more like "inner mind": particularly thoughts which seem to come into your mind spontaneously, which Elves think come from the Valar or even from God. Humans, think the Elves, are bad at listening to their ore; it may even be that their ore gives them bad advise, because of the actions of Morgoth/Satan. Tolkien the linguist is cares that Elvish words do not correspond exactly to English words; Tolkien the philosopher cares that Elves have their own perception of how psychology and conscience operate; Tolkien the Catholic cares about the consequences of original sin in his sub-creation. We're a very long way from translating Hamlet back into the original Klingon.
Hostetter helpfully notes that when Tolkien said that the Lord of the Rings is fundamentally Catholic, he does not mean that it is explicitly or directly Catholic (in the way that C.S Lewis's fairy tales and science fiction stories were directly and explicitly Christian): he means literally that they have a Catholic foundation. The building blocks of the world take Catholic thinking for granted. It is clear that, in a similar way, the Lord of the Rings is fundamentally linguistic. Elves have a particular perception of time; and a particular way of thinking about the future; and a particular view of conscience because Tolkien was concerned about the nuances of the word gor; and a thousand other words as well.
There is a kind of writer who cannot describe a character opening a door unless he first decides what the doorknob looks like. Perhaps we need to think of Tolkien as an historical novelist, writing stories in an imaginary history which he had created himself, but a history nonetheless. He could no more make up a fact about Elrond than Hilary Mantel can make up a fact about Thomas Moore. And his history is created out of philology and Thomist theology: in asking these incredibly dry, abstract questions, he is bringing the world into sharp enough focus that he is capable of writing about it. If Boromir had told Frodo that his favourite food was an orc-plant, Tolkien would have believed it because he hadn't, in that sense, just made it up.
So it seems that that was how Tolkien wrote. How Middle-earth grew; how it congealed on the page. If you love the Lord of the Rings and the Silmarillion then you have to say that this was the right thing to do; a different writer would have done it differently and we would have ended up with a different book.
Still: the Nature of Middle-earth is a sad book. Middle-earth, so solid, so convincing, "a world more real than any other", dissolves into a series of calculations and conjectures; an unknowable, unfinishable work-in progress. Sad because Tolkien was so obviously struggling to do the impossible. An old man, working out his backstory to three hundred significant figures, when we would rather he had written one more poem or a single footnote about the Entwives -- when we would much rather he had not abandoned his "thriller" about Minas Tirith during the reign of Aragorn's son.
And because it seems like such a waste.
On one occasion the BBC declared the Lord of the Rings the 26th greatest book of the twentieth century; on another, the best loved British novel of all time. It is estimated to have sold a hundred and fifty million copies, which would make it comfortably the best selling "authored" book of all time.
If Tolkien hadn't wasted his time worrying about pedantic detail, imagine what he could have achieved.