On Fairy-stories was composed over the critical period (1939-1943) in which Tolkien wrought the One Ring out of the magic ring.
The fundamental idea of the essay is that fantasy art is a ‘magical’, sub-creative act. The true fantasy art is elvish enchantment, mortal artists merely aspire to it, while (reading between the lines) the forging of the One Ring involves the same operations but is a perversion because desire for power has corrupted the will to enchant.
Immediately after giving an elvish definition of fantasy, Tolkien redefines his earlier terms. Magic, used hitherto as synonymous with enchantment, is reclassified as (in effect) necromancy. This redefinition provides a pivot from The Hobbit to The Lord of the Rings: an originally innocent magic ring becomes altogether evil, without changing its name.
The newly devised opposition between necromancy and enchantment frames the new hobbit story, in which art (symbolized by the tower looking over the sea of Frodo’s dream at Crickhollow) defeats magic (the One Ring that is the foundation of the Dark Tower).
But this pivot from The Hobbit to The Lord of the Rings, from magic to magic vs. enchantment, is accompanied by a generalization of the original magic ring into what Tolkien calls fairy-elements, the basic ingredients of fairy-stories.
Tolkien (in the section ‘Origins’) presents fairy-elements linguistically: noun-adjective combinations, rarely found in the world. Comparing early draft (1930) and published (1937) second sentence of The Hobbit suggests hobbit’s hole (1930) inspired an early fairy element: hobbit-hole (1937); but (as G.R.R. Martin seems to have grasped) the original fairy-element of The Hobbit was the hole-door marked by a rune on the map of Thror. In his essay, Tolkien gives other instances of fairy-elements: swan-robes, wicked-stepmothers, arbitrary-prohibitions, a detachable-heart, the elves themselves (elvish-artists), and magic-rings.
Tolkien’s account of fairy-elements as noun-adjective combinations reworks older Oxford theories of naming into a linguistic theory of the nameless. He signals his intent in the introductory section of the essay, in which fairy-stories are observed stories about the realm of Fairie, which land, he insists, alludes definition. In other words, a defining quality of Fairie is that, the best efforts of storytellers notwithstanding, it resists naming. By implication, a fairy-element, which appears to name the denizens and things of Fairie, is not a name, at least not by dictionary standards.
A fairy-element is cast as a metaphor, picturing what can be seen but not named.
Tolkien’s theory of fairy-elements develops the magic ring of The Hobbit, a sign of the nameless self. The adventure of Bilbo Baggins spins the meaning of this magic ring by weaving that of his proper name, out of an original sentence that appeared out of nowhere: In a hole in the ground there lived a hobbit. The mysterious sign at the center of the story, a name of the nameless, takes its meaning from the story – for, Tolkien knew, only a story gives meaning to a name. But this sign of the nameless self is a name only to an Ent, usually a story is reduced to a single word or word-combination, which serves as a name but may be a fairy-element.
On Fairy-stories generalizes the magic ring into the fairy-element, explored from the perspective of storytelling. A spell of enchantment works as the story told makes a fairy-element credible, and the audience begin to take literally those word-combination pictures that are, as it were by definition, metaphors.
Because this art of fantasy bewitches the mind those operations that replicate it for the purpose of dominating other minds are black magic.
* * *
The upshot of all this is that On Fairy-stories marks the passage by which the magic ring at the center of the original story of Bilbo Baggins begat both the One Ring and the tower looking over the sea, the symbol of the lost art of northern story.