Magic, Wonder, and Mystery

Introduction

Steven Erikson's approach to magic in the Malazan Book of the Fallen runs directly against the dominant trend of twenty-first-century epic fantasy, which has embraced increasingly systematised, rule-bound, and rationalised magical apparatuses whose limits and costs are specified to the reader in sufficient detail that the reader can predict what any given spell will do. Writers like Brandon Sanderson have made this approach — often described as "hard magic" — a central craft principle and have articulated it through widely circulated maxims (notably "Sanderson's First Law": an author's ability to solve conflict with magic is directly proportional to how well the reader understands said magic). Erikson's position is the polar opposite. He has argued, in multiple recorded interviews and essays, that fully explaining a magical system drains it of its defining property — its capacity to evoke wonder — and that the correct craft response is to maintain the rules as authorial constraints without ever surfacing them to the reader as reader information.

This essay reconstructs Erikson's theory of magic under six headings: the diagnosis of what magic is for within the form of fantasy; the inverse-Clarke principle (any sufficiently described technology is no longer magic); the distinction between authorial rules and reader rules; the function of wonder as a connection to childhood imagination; the Anomander Rake case study in which a character's unknowable depth becomes the template for treating more ordinary characters at a distance; and the epistemic consequences of the approach for the reader, who must be trained to accept that not-knowing is the intended state rather than an obstacle to be overcome.

What Magic Is For: The Evocation of Wonder

Erikson's theoretical starting point is the question of function. Before deciding how magic should work in a fantasy novel, the writer should ask what magic is for — what purpose its presence serves within the fictional world and the reading experience. Erikson's answer, articulated most directly in Steven Erikson Talks Building Malazan, grounds the question in the genealogy of fantasy itself:

"The magic system requires one to step back and think of the origins of fantasy as a genre, and those origins lie in mythology and legends and other stories. And then you look at what the role of magic is in those stories, and I would define the role of magic in those stories as the evocation of wonder and mystery. And so if that is the purpose for, you know, swords and stone and ladies in the lake and all the rest — then why on earth would I want to try to apply a... view magic as an engineering solution, or as a scientific problem, or as a technical issue, and then build that up more or less no different from any technology? That to me defeats the purpose of magic in fiction. So it has to stay mysterious, it has to evoke a sense of wonder. We certainly follow specific rules, we have limits, but we certainly do not at any time feel obliged to explain them, because that kind of defeats the purpose." (Steven Erikson Talks Building Malazan, Facebook Post & More transcript)

Three moves in this passage are load-bearing. First, Erikson grounds the question in a historical observation about where fantasy came from. The genre's ancestors — the Arthurian cycle, Norse mythology, the Homeric epics, the Grimm folk-tales — treat magic as encounter with the inexplicable, not as operation of a legible rule-system. The Lady of the Lake does not explain how she enchanted Excalibur; Circe does not document the mechanism by which her wand transforms men into swine; the Norns do not publish their metrics. Magic in the founding works of the genre is precisely the thing that cannot be fully explained, and its capacity to move the reader is proportional to its resistance to explanation. Any reduction of magic to rule-system is therefore a departure from the historical function of magic in the form, and any defence of the reduction must acknowledge that it is exchanging one kind of aesthetic effect for another.

Second, the passage makes explicit the choice between two alternative craft paths. A writer can treat magic as technology — something whose properties are specifiable, whose limits are quantifiable, and whose deployment in the plot can be checked for consistency — or the writer can treat magic as mystery — something whose properties evade specification, whose limits are suggested rather than stated, and whose deployment in the plot operates under constraints the author knows but the reader is not permitted to see. Both paths are legitimate craft choices, but they produce radically different reading experiences. The technology path produces a reader who can predict outcomes and who derives pleasure from the satisfaction of seeing a rule-consistent system operate; the mystery path produces a reader who cannot predict outcomes and who derives pleasure from the experience of continually encountering something irreducible to their previous categories.

Third, the passage admits the existence of rules while refusing to publish them. "We certainly follow specific rules, we have limits, but we certainly do not at any time feel obliged to explain them." The distinction is crucial: Erikson is not advocating arbitrary magic in the sense of magic that could do anything at any moment, because arbitrary magic would have no dramatic weight (nothing would matter, because anything could be reversed). He is advocating constrained but undisclosed magic — magic that operates under rules the author has accepted but whose rules the reader must infer from observation of the magic's behaviour rather than from exposition. The effect is a fiction that feels mysterious without feeling incoherent, because the mystery is produced by the reader's incomplete access to a system that is internally consistent.

The Inverse-Clarke Principle

A useful way of formalising Erikson's position is as the inverse of Arthur C. Clarke's famous dictum that "any sufficiently advanced technology is indistinguishable from magic." Clarke's principle describes the epistemic condition of observers encountering technology whose operating principles exceed their theoretical understanding: the technology's effects appear magical because the observer cannot construct the causal chain that would demystify them. The inverse — any sufficiently described technology is no longer magic — describes the craft condition of writers who have pushed a magical system too far in the direction of systematisation: once the system has been described in sufficient detail, it ceases to function as magic and becomes instead a fictional technology with a different name.

Erikson's position is that the inverse-Clarke principle is the craft problem contemporary hard-magic fantasy has refused to take seriously. A writer who has produced a set of rigorous rules for a magical system has, by the inverse principle, produced a technology. The prose may still use the word "magic," but the phenomenological effect on the reader is the same as the phenomenological effect of reading a science-fiction novel whose engineering apparatus has been laid out in detail. The reader experiences the system as operable, calculable, and (eventually) predictable, and the sense of encountering something mysterious has been traded away for the sense of encountering something coherent.

Erikson's refusal of this trade-off is not a rejection of coherence but a rejection of the assumption that coherence and mystery are incompatible. His practice in the Malazan series demonstrates that a magical system can be both coherent (the same character's abilities behave consistently across volumes; warrens have identifiable properties that distinguish them; specific kinds of magic have recognisable costs) and mysterious (the reader cannot predict what a given mage will do in a given situation, cannot extract a complete rule-set from the prose, and continues to encounter new aspects of the system even in the final volumes). The coherence is the author's; the mystery is the reader's; and the two are not in conflict because they operate on different sides of the prose.

Authorial Rules vs. Reader Rules

The analytic key to Erikson's method is the distinction between the rules the author operates under and the rules the reader is permitted to see. Every serious writer of secondary-world fiction has authorial rules — internal consistencies they have established for themselves about what the world can and cannot do — but writers differ widely in whether they surface those rules to the reader. A hard-magic writer like Sanderson surfaces the rules explicitly, often in dedicated expositional passages that function as the reader's tutorial in how to think about the magic. A soft-magic writer like Tolkien keeps the rules largely implicit, surfacing only the broadest outlines and leaving the detail to the reader's inference. Erikson's practice sits at the extreme end of the Tolkien pole: the authorial rules exist, but their surfacing to the reader is minimised to the point where the reader's understanding of the magic is always partial and always constructed from indirect evidence.

The craft principle is that partial understanding is the phenomenological precondition of wonder. A reader who fully understands the magic cannot experience it as magical, because the experience of magic requires the sense that the phenomenon exceeds one's conceptual grasp. A reader whose understanding is deliberately kept incomplete experiences the magic as an ongoing encounter with something not-yet-comprehended, and the not-yet-comprehended is the form wonder takes. Erikson's prose is therefore calibrated to keep the reader in this state — to provide enough evidence that the magic is doing something specific and consequential, but never enough evidence to let the reader reconstruct the complete rule-set. The calibration is what distinguishes his practice from writers who are merely vague; vagueness would not work, because vagueness leaves the reader with nothing to attach to, whereas Erikson's partial specificity gives the reader plenty to attach to while preserving the shortfall that makes the attachment generative.

Wonder and Childhood: The Case for Preservation

Erikson has connected his insistence on preserving magic's mystery to a broader observation about the function of wonder in human development. Wonder, in his account, is not an ornamental emotion that accompanies aesthetic experience but a specific cognitive capacity most evident in childhood and progressively eroded by the standard operations of adult rationalism. Children experience the world as saturated with possibility because their categorical systems are still incomplete; any new phenomenon may turn out to be something their existing categories cannot accommodate, and the possibility of being surprised is always present. Adults, having acquired categorical systems sufficient to classify most encountered phenomena, experience the world as progressively less surprising, and the progressive loss of surprise is (in Erikson's implicit account) one of the costs of adulthood that fantasy fiction is uniquely positioned to remedy.

The remedy works by providing adult readers with encounters that cannot be absorbed into their existing categorical systems. A magical system whose rules have been fully explained can be absorbed immediately — it becomes one more technology to be catalogued — and therefore offers the reader no recovery of the childhood experience of surprise. A magical system whose rules are deliberately kept beyond the reader's full grasp cannot be absorbed; it sits in the reader's cognitive apparatus as a permanent excess that the reader must continue to engage with rather than mastering and filing away. The reading experience of such a system is structurally similar to the childhood experience of encountering the genuinely unfamiliar, and the pleasure is the pleasure of being, briefly, a cognitive agent for whom the world has not yet been exhausted.

This is a demanding theoretical claim and it rests on an aesthetic preference Erikson shares with writers like Ursula K. Le Guin and Tolkien himself, who both articulated versions of the same position. Tolkien's essay "On Fairy-Stories" (1947) argues that fantasy's principal function is the "recovery" of what he called the "clear view" — the restoration of wonder at ordinary things by the act of placing them in an extraordinary context. Le Guin's "From Elfland to Poughkeepsie" (1973) argues that the dilution of fantasy's stylistic register into modern colloquial prose destroys the genre's capacity to evoke the sense of encounter with the marvellous. Erikson's position is in this tradition: magic's purpose is to preserve wonder, and the craft decisions that preserve wonder are the craft decisions that define good magical writing.

Anomander Rake: The Character as Mystery

The Malazan series' most sustained application of the magic-as-wonder principle at the level of character is Anomander Rake — the Son of Darkness, Knight of High House Dark, Lord of the Tiste Andii, whose age is measured in hundreds of thousands of years and whose interior life the novels deliberately refuse to render directly. Erikson has discussed the craft decision to hold Rake at a distance:

"Anomander Rake doesn't remind me of an Aragorn. He reminds me more of like a walking Achilles — like he's a walking myth among men... I tried, I mean I tried one early draft of Gardens of the Moon, I realised I can't do this. This character is 200,000 years old or whatever, right? And there's no way I can match or even attempt to establish any verisimilitude in terms of his point of view or his thoughts. So I would hold him at a distance. That was a good lesson, because I then decided to start holding other characters who are not nearly as powerful at a distance — [an] extreme one of that would be Tavore. And holding her at a distance was one of the great joys of the writing of the series. I absolutely adored doing that — to keep her mysterious all the way through until... there's one scene where we get inside. Until then, and of course everything was leading up to that, so the more cagey I got the happier I was until I reached that one moment of release." (Steven Erikson Talks Building Malazan transcript)

Three observations follow from this disclosure. First, the decision to hold Rake at a distance was originally a pragmatic choice forced by the impossibility of rendering a 200,000-year-old consciousness from the inside. Erikson tried and failed, and the failure taught him that some characters are structurally incompatible with conventional point-of-view technique. The pragmatic failure produced a craft insight — that holding a character at a distance could become a method rather than a limitation — and the method was subsequently generalised.

Second, the generalisation is what matters. Erikson extends the distancing technique from characters who must be distanced (because their cognitive scale exceeds the reader's) to characters who could be rendered from the inside but who benefit from being held at a distance anyway. Tavore is the paradigm case. She is a contemporary human whose interior monologue could in principle be written at any time, but Erikson's decision to keep her at a distance for nearly the entire series — surfacing only rare glimpses of her interior, and saving one sustained access moment for near the end — produces a character whose mystery is the driving force of her arc. Readers spend volumes trying to construct Tavore from external evidence, and the construction is the reader's activity in a way that a fully accessed interior would have foreclosed.

Third, the Rake/Tavore principle extends the magic-as-mystery logic from the magical apparatus to character construction itself. A character whose interior is deliberately withheld is, in the same structural sense as a magic system whose rules are withheld, a source of wonder. The reader's incomplete access produces the felt sense that there is more to the character than the prose has surrendered, and the felt sense is what makes the character compelling. When the surrender finally comes — when Tavore's interior is briefly opened — the release is earned by the thousands of pages of withholding that preceded it, and the release's emotional force is directly proportional to the discipline of the withholding.

The Reader's Training: Accepting Not-Knowing as the Goal

The practical consequence of Erikson's approach is that readers of the Malazan series must be trained, over the course of the volumes, to accept not-knowing as the intended state rather than a problem to be solved. The training is one of the most distinctive features of the reading experience, and it is one of the reasons why some readers find the series frustrating in its early volumes and captivating in its later ones. Readers who arrive with expectations shaped by hard-magic contemporaries (Sanderson, Rothfuss's Kvothe, the rule-set fantasies of the last two decades) often spend the first few hundred pages attempting to extract a rule-system from the prose and growing frustrated that the system refuses to disclose itself. Readers who eventually adjust — who recognise that the rule-system is deliberately withheld and that the withholding is not a failure but the craft — are able to enter the experience the prose is actually offering.

The adjustment is cognitive as well as aesthetic. A reader trained by hard-magic fantasy has been conditioned to treat incomplete information as a gap to be filled, a puzzle whose missing pieces must be identified and inserted. A reader trained by Erikson's prose learns instead to treat incomplete information as a condition to be inhabited — a state of ongoing partial comprehension in which the incompleteness is not a problem but the substance of what is being offered. The shift from puzzle-solving to inhabitation is the cognitive equivalent of the shift from technology to wonder, and the shift is what the series is (in part) in the business of training. Readers who complete the ten volumes have been reshaped as readers, and part of the reshaping is the acquisition of the capacity to experience mystery without trying to dissolve it.

This is why re-reading Erikson is different from re-reading many hard-magic contemporaries. A re-reader of Sanderson discovers that the puzzles they could not solve on the first pass are now solvable, because they now possess the rule-set. A re-reader of Erikson discovers that the aspects they could not understand on the first pass are still incompletely understood, but the incomplete understanding has become richer — new connections become visible, new registers come online, and the mystery deepens rather than dissolving. The magic has preserved its wonder-generating property across the re-read because the craft decisions that produced the mystery on the first pass were not puzzles whose solutions could be acquired; they were calibrated withholdings whose withheld content was never going to be fully delivered.

Conclusion: Wonder as Craft Commitment

Erikson's theory of magic is, in the final accounting, a defence of wonder as a legitimate aesthetic target and a craft claim about the specific techniques that produce it. The theory rests on three commitments. First, that fantasy's historical function is the evocation of mystery and the re-creation of a childhood-adjacent capacity for surprise, and that any practice which systematically dissolves mystery (however rigorous and internally consistent) is working against the genre's formative purpose. Second, that coherence and mystery are not incompatible: a magical system can be internally consistent at the level of authorial rules while remaining mysterious at the level of reader access, and the distinction between authorial rules and reader rules is the craft distinction that makes the combination possible. Third, that the distancing technique applicable to magic is equally applicable to character: holding a character at sufficient distance that their interior is never fully rendered produces the same wonder-generating effect as holding a magical system at sufficient distance that its rules are never fully disclosed.

The combination of these commitments produces a fiction that is harder to read than most contemporary fantasy in the short term and more durably rewarding in the long term. The difficulty is the cost of the calibration: readers who want to be told what is happening will not be told, and readers who want to master a rule-set will not be given one to master. The reward is the experience the calibration produces: a reading in which wonder is never exhausted, because the craft has been designed to keep it from being exhausted, and a re-reading in which the same wonder deepens rather than evaporating. This is what Erikson has meant by his insistence that magic must stay mysterious — and the Malazan series is the ten-volume demonstration that the insistence produces, when followed to its craft conclusions, fiction of a specific and distinctive kind that the hard-magic alternatives cannot match.

The broader implication for the genre is that the question "how should magic work in a fantasy novel?" is the wrong question. The right question is "what effect do I want magic to have on my reader, and what craft decisions produce that effect?" The hard-magic answer — coherent predictability satisfying to the systematising imagination — is one possible answer, and it has produced good fiction in the hands of skilled practitioners. Erikson's answer — partial specificity producing sustained wonder — is another, and it has produced fiction whose effect on attentive readers is, in the phrase that keeps recurring in reader testimony, transformative. Neither answer invalidates the other. But the two answers produce different books, and the readers who want what Erikson is offering will not find it in the writers who have chosen the other path.

Sources

Related Essays