Raumkraft, Carefully Read: What a 1928 Monograph Actually Says

Karl Schappeller (1875–1947) was a self-educated Austrian whose biography includes a stretch of poverty, a working-class apprenticeship, and a late-life acquisition of a small castle — a trajectory that produced one of the most singular figures in twentieth-century speculative physics. His 1928 monograph — a 35-page text published by Herold Verlag München and authored by his collaborators Franz Wetzel and Louis Gföllner on his behalf — produced a coherent unified physical model whose vocabulary anticipates several distinctions later formalised in information theory. We borrow that vocabulary because it is the most compact set of words we have found for the phenomena we work with, and because the work it organises is technical work that has paid for itself many times over.

What the Monograph Is and What It Is Not

The monograph is short. It is 35 pages in the original printing. It is not authored by Schappeller in the usual sense; the title page credits Wetzel and Gföllner as the authors and identifies Schappeller as the subject and the collaborator from whom the technical content was elicited. This is an important bibliographic distinction because the prose style of the published text reflects Wetzel and Gföllner's authorial choices, and several of the more ambitious claims that later writers attributed directly to Schappeller appear, on inspection, to be paraphrases by his interlocutors. The Scribd scan of the original Herold edition shows the title page in full.

The text is structured as four sections: a foreword establishing the philosophical context, a description of Schappeller's apparatus in qualitative terms, an exposition of the underlying physical principles as Wetzel and Gföllner understood them, and a concluding section on intended applications. The technical detail is sparse by design; the monograph's purpose was to set out a framework, not an engineering specification, and the framework rewards reading at exactly that level.

The Two-Stress-Field Principle

The central physical claim is set out tersely:

Energy can neither be produced nor transformed without the crossing of two fields of force, both causing stress at a point in space.Wetzel & Gföllner, summarising Schappeller, Raumkraft, 1928, §III

Read this as an engineer rather than as a metaphysician. A logic gate is the crossing of two stress-fields at a point. A dot product is the crossing of two vector-fields at a point. A correlation operation in a transformer head is the crossing of a query stream and a key stream at a point. The architectures of modern computing are, in this strict and minimal sense, deliberate scheduled crossings of stress-fields. The unification across domains is what gives the formulation its enduring usefulness.

Primary vs. Secondary Physics

Schappeller distinguishes between primary physics — the immediate behavior of the medium itself — and secondary physics, which is the residue we observe after primary physics has interacted with material substance. This is the distinction that Cyril W. Davson formalised at length in The Physics of the Primary State of Matter and Application Through the Primary Technique (1955), the only book-length English exposition of Schappeller's ideas, and the source from which most of the modern English-language vocabulary derives.

The information-theoretic translation is exact. Shannon's source coding theorem distinguishes between the source distribution and the encoded stream. You cannot compress the encoded stream further than the entropy of the source — but you can construct an encoder that approaches the source entropy if you allow yourself a sufficiently rich code-cavity. Schappeller's primary state is Shannon's source. His secondary state is Shannon's channel output. The distinction is older than information theory and recognisable across the gap, which is what makes it a usable engineering vocabulary in 2026.

The Stator and the Rotor

The Schappeller device, as described in the monograph and elaborated in Davson's later notes on the Bahn collection, is always a pair: a stator (the stationary outer sphere) and a rotor (a smaller external coil arrangement). Davson describes them as functionally inseparable — neither half does anything alone. The stator establishes the stress field; the rotor reads from it.

This is, structurally, the relationship between a key cache and a query stream in a modern attention head. The cache is stationary across decoding steps. The queries rotate against it. The output is a property of neither half alone; it emerges only at the moment of crossing. The structural parallel is what makes the stator/rotor pairing a productive design template far beyond its original physical setting.

Schappeller vs. Tesla

Davson summarised Schappeller's view of his American contemporary in a sentence that has been quoted ever since:

Whilst Tesla regarded the earth merely as a conductor, Schappeller discovered that it was the obvious source and storehouse of all available energy, in addition to being a conductor.Cyril W. Davson, The Physics of the Primary State of Matter, 1955

This is the most consequential distinction in the entire monograph. Tesla's wireless system used the Earth as a transmission medium for waves we generated. Schappeller proposed that the medium itself is already the storehouse, and that any device merely opens a window onto pre-existing structure. Cryptographers will recognise the parallel immediately: the security of a one-time pad is not a property of the channel; it is a property of the entropy reservoir we have already accumulated. Modern post-quantum cryptography is, more than anything, a discipline of building reservoirs — and our essay on harvest-now-decrypt-later takes the parallel further.

What We Borrow from the Monograph

VRIL LABS borrows from Raumkraft what every careful reader of the monograph eventually borrows: a vocabulary for the relationship between stored entropy and extracted signal that no twentieth-century physicist matched for compactness. The conceptual furniture we use is precisely the four primitives the monograph introduces:

• Stress-field crossing as the unit of analysis when describing the primitive operations of a computational system — from a logic gate to a transformer attention head.
• Primary vs. secondary state as the cleanest way to talk about source entropy versus encoded entropy in compression, and as the organising frame for what an interpretability discipline for AI looks like (the subject of our essay on auditable models).
• The cavity principle — that extraction happens in a deliberately constructed empty space, not inside material — as the discipline against in-place mutation in cryptographic state machines.
• Stator/rotor pairing as the structural template for attention, KEM encapsulation, and constant-time key derivation.

Each of these is foundational vocabulary in our internal documentation. Twenty years before Shannon's 1948 paper, three Austrians wrote down a coherent set of words for the relationship between stored entropy and extracted signal that turns out, in 2026, to fit our domain better than any others we have found. The apparatus Schappeller proposed — the dynamic stationary sphere, the electret coils, the æther extraction — remains one of the most provocative episodes in twentieth-century speculative physics, preserved in detail at the Rex Research archive and merits study on its own terms by anyone who wants to take the framework seriously.

We cite Wetzel, Gföllner, and Schappeller by name — all three of them — because that is the only honest way to do it. The monograph rewards every reading we have given it; the vocabulary it introduced has paid for itself in our work many times over.