On Magic, and the Things It Does Not Do

The difficulty in writing seriously about magic is that nearly everyone already believes himself familiar with the subject. A child has heard stories of sorcerers turning armies into crows, witches cursing entire bloodlines with a glance, kings made immortal by enchanted crowns, and towers raised from empty fields before breakfast. By adulthood he has usually learned that these are stories, though he generally retains the less obvious assumption underlying them: that real magic is simply the same phenomenon practised on a smaller scale. It is not.

Magic exists. It can be observed, measured, taught, repeated, and, within limits, relied upon. A trained practitioner may produce heat without flame, influence the motion of small objects, alter sensation, detect certain substances at a distance, accelerate some chemical processes, interfere with others, produce limited barriers against force, and accomplish several dozen additional feats which would unquestionably appear miraculous to a person unfamiliar with them. What he cannot ordinarily do is whatever the storyteller requires in the third act.

No competent mage can wave his hands over a field and produce a castle. No healer restores a severed head because he has studied particularly hard, and no known spell transforms an army into swine, reverses yesterday, makes a man permanently invulnerable, or summons enough bread from nothing to solve a famine. The great majority of practical magic is bounded by distance, duration, concentration, complexity, and the physical circumstances under which it is performed. This does not make it unimpressive.

A flame no larger than one’s thumb is unimpressive until it is produced inside a sealed boiler housing where no ordinary flame can safely be introduced. The ability to move an object weighing four ounces is not particularly dramatic until that object is a surgical instrument inside a wound. A spell which detects traces of arsenic in water will never inspire an epic poem; it has nevertheless saved more lives than several heroes who have. Civilisation is constructed largely from unimpressive things applied correctly.

There is also no meaningful opposition between magic and natural law, despite several schools of mysticism insisting otherwise. Magic does not occur outside nature; if it did, we could not study it. A levitated object has not escaped physics, and something has caused it to remain where gravity says it should not. Discovering the nature of that something is precisely the physicist’s problem. Calling an unexplained mechanism “magic” does not exempt us from asking how it works. It merely tells us which department ought to receive the grant.

Most magical practice is therefore considerably less romantic than the public imagines. Professional mages spend a surprising amount of time preparing materials, repeating exercises, calibrating instruments, maintaining concentration, and correcting mistakes made by apprentices who believed that enthusiasm could substitute for precision. Industrial thaumaturges frequently work beside chemists and engineers, and medical practitioners consult anatomists. Military mages spend rather more time discussing sight lines, weather, fatigue, and logistics than hurling fire at anyone. The wizard standing alone upon a hill commanding the elements is a splendid image; the wizard employed by the municipal water authority checking twenty-six warded pressure valves is a more representative one.

There is, however, one branch of the magical arts for which my preference for modest description becomes considerably harder to maintain. Runecarving is ancient, poorly understood, exceedingly difficult, and, if the surviving evidence is interpreted correctly, capable of effects so broad that one must be very careful before declaring anything categorically impossible.

The basic principle is straightforward enough to state. Certain symbols, when carved with sufficient precision into an object and inlaid with particular materials, alter the physical behaviour of that object. The shape of the rune matters, as do its dimensions, the material into which it is cut, and the substance used for the inlay. Orientation frequently matters, and so does the relationship between one rune and another. Change any of these carelessly and the effect may weaken, disappear, or become something the carver did not intend. This is why one should resist the popular description of runes as a magical alphabet: they are no more an alphabet than the teeth of a gear are handwriting.

The simplest examples are well understood. Several common configurations reduce apparent weight, and a properly carved carrying rune can make an iron chest manageable by one person while leaving its hardness and dimensions largely unchanged. More sophisticated arrangements reduce effective weight nearly to nothing, or produce stable levitation when combined with appropriate anchoring patterns. The immediate question is inevitably why we do not simply carve such runes onto every ship, locomotive, bridge component, artillery piece, warehouse crate, and inconveniently heavy piano. The answer is that people have tried.

Runes wear, and materials interfere with one another. A large object requires a proportionately more complex array. Some configurations become unstable under vibration or heat; others require expensive inlays, or interact unpleasantly with nearby magical fields. A pattern which works beautifully upon a laboratory plate may fail when scaled tenfold, because the geometric relationships are not merely decorative. The universe is disappointingly resistant to convenient manufacture.

True runemasters, however, work beyond these elementary forms. A complex runic array is not simply a collection of individual symbols, each producing a separate effect. The runes interact, and their positions relative to one another appear to create higher-order behaviour which cannot be predicted merely by adding their individual properties together. This is where the subject ceases to be comfortable.

We possess surviving examples of runic work which regulate temperature, redirect force, suppress vibration, maintain pressure, preserve organic material, and alter the transmission of sound. A small number appear to respond differently under different physical conditions, suggesting that the array itself incorporates something resembling conditional behaviour. I use the word resembling deliberately, because every few years some enthusiast declares that sufficiently advanced runes can “think.” They cannot. At least, I very much hope they cannot.

What is undeniable is that complexity increases capability at a rate which has frustrated every attempt to establish a firm theoretical ceiling. An accomplished carver may combine dozens or hundreds of elements into a single geometric system, and surviving ancient works contain arrays of thousands. At that scale the practical limitations cease to be primarily magical and become mathematical, material, and spatial. The carver requires enough surface upon which to place the design, materials pure enough to preserve its behaviour, and an understanding of the relationships between components sufficient to avoid turning several years of work into expensive decoration.

This has inspired another popular misconception: that any magical effect can be achieved through runes if only the pattern is sufficiently complicated. It has not been demonstrated. It is merely difficult to prove false, and there is an important difference.

No living runecarver can produce a mountain from empty air, and no authenticated ancient array does so either. Yet one can readily imagine an array which makes stone lighter, redirects structural load, reduces friction during transport, and stabilises a mass during placement. To an observer watching a properly equipped construction crew move a hundred-ton block with a dozen labourers, the distinction between “created a mountain” and “made mountain-building absurdly easier” may feel unnecessarily scholarly. I assure the reader that distinctions of this kind are why civilisation continues to possess engineers.

Runecarving also differs from ordinary spellcraft in one philosophically important respect: once properly made, the object does the work. A mage becomes tired; a rune does not appear to. A spell ends when concentration fails or whatever sustaining condition it requires is removed, whereas a runic object may continue behaving according to its altered properties for decades or centuries, provided the pattern remains intact. Some ancient works are still functioning, which is among the more unsettling facts in the field.

We possess devices whose makers are unknown, whose exact purpose is disputed, and whose inscriptions continue doing something after kingdoms, languages, and entire systems of religion have disappeared around them. Modern scholars can often identify individual elements within these arrays while remaining unable to explain the whole. It is rather like discovering an operating locomotive built by a civilisation which left us no instructions, then realising that our best engineers understand the wheels but not the engine. Naturally, several museums display such objects. I have expressed reservations; museums have ignored me, as is customary.

One must therefore maintain two thoughts about magic simultaneously. The first is that almost everything commonly said about it is exaggerated. Real mages are constrained practitioners working within reproducible limits rather than minor gods who have inexplicably chosen academic employment, and most magical work is useful precisely because it is modest, reliable, and applied to problems which would otherwise require more labour, more machinery, or greater risk.

The second is that our present limitations should not be mistaken for the limitations of magic itself. Runecarving alone provides sufficient evidence that combinations of comparatively simple magical effects can produce consequences of extraordinary sophistication, and whether there is an ultimate boundary to that sophistication remains unknown. I dislike unknown boundaries, which encourage poets, but intellectual honesty requires admitting that we do not know where this one lies. The fairy tales are wrong when they tell us that magic can do anything; the more interesting possibility is that they are wrong only about how much work it takes.