Blackmoor — Spatial Archaeology Before Dungeons & Dragons¶

1. Purpose of the research¶

This research was indirectly born during the development of Lands of Folklore.

While implementing diagonal geometry in a discrete-movement dungeon crawler, a seemingly simple question arose:

How did earlier RPGs and dungeon crawlers solve the problem of representing spaces that don't fit cleanly into an orthogonal grid?

Initial research on classic CRPGs led to tabletop RPGs and, finally, to Blackmoor, the campaign started by Dave Arneson around 1971.

Early Blackmoor maps contain numerous diagonal corridors, irregular rooms, oblique intersections, and a complex vertical structure.

This raised a second, much more interesting question:

What did the grid really mean to Arneson and his players?

The hypothesis that has emerged is that we are projecting onto those maps a modern conception of the grid that they may not have had.

2. Context: Blackmoor did not appear as a finished RPG¶

DOCUMENTED FACT

Blackmoor emerged within the Twin Cities wargaming community and the experimental environment generated around David Wesley's Braunstein.

Braunstein had introduced fundamental elements:

participants controlling individual characters;
different objectives for each participant;
incomplete information;
actions not necessarily anticipated by the rules;
a referee in charge of adjudicating their consequences.

Arneson participated in this environment.

There is also contemporary documentation from April 1971 announcing Arneson's "medieval Braunstein", an immediate antecedent of Blackmoor.

The exact chronology of the first sessions remains under investigation.

Some sources and reconstructions place campaign antecedents as early as 1970, while during 1971 we find clearer evidence of Blackmoor's development.

CONCLUSION

Blackmoor does not seem to have appeared through prior writing of a new ruleset.

It seems to have evolved through play.

3. A campaign that generated its own rules¶

One of the most important characteristics of Blackmoor is that it is difficult to separate:

the game

from

the process by which Arneson was inventing it.

Arneson used elements from different systems and modified them when they stopped solving problems arising during play sessions.

A particularly significant example is Chainmail.

POSTERIOR TESTIMONY OF ARNESON

Arneson later explained that initially they used Chainmail, but that his system was fundamentally designed for military confrontations and didn't work satisfactorily when the focus shifted to individual characters.

According to his recollection, his effective use lasted approximately a couple of sessions before being replaced.

This questions the simplified genealogy:

Chainmail
   ↓
Blackmoor
   ↓
D&D

A more adequate description could be:

Wargaming
   │
Braunstein
   │
Chainmail ─────┐
               │
Ironclads ─────┼──► BLACKMOOR
               │       │
own ideas ─────┘       │
                       ▼
               experimentation
                       │
                modification
                       │
                       ▼
                 new rules

Blackmoor therefore seems less like a ruleset that became a campaign and more like a campaign that constantly required producing new rules.

4. Civil War Ironclads and the birth of new abstractions¶

Arneson later claimed that some concepts used to resolve individual combat came from a system he had previously used for Civil War naval battles.

In that context existed equivalents to:

protection;
ability to withstand damage;
degradation before destruction.

This provides a plausible connection between Civil War Ironclads and certain abstractions that would later be recognized as Armor Class and Hit Points.

Evidence status¶

DOCUMENTED

Arneson declared this influence.

NOT YET DEMONSTRATED

the exact mathematical genealogy between the rules of Ironclads, the early rules of Blackmoor, and the later ones of D&D.

Therefore, it should not be written:

"Arneson took AC and HP directly from Ironclads."

The prudent formulation would be:

Arneson later identified his experience with Civil War Ironclads rules as one of the sources used to develop protection and damage-resistance systems appropriate for individual combat.

The exact evolution remains pending research.

5. Blackmoor was much more than a dungeon¶

Another historiographical pitfall is to reduce Blackmoor to:

"the place where the dungeon crawl was invented."

Blackmoor was a campaign.

It included different scales:

individual characters;
exploration;
adventures;
settlements;
politics;
armies;
economy;
territories;
construction;
military campaigns;
dungeons.

Some of this material was later preserved in The First Fantasy Campaign (Judges Guild, 1977).

This allows understanding a peculiarity that D&D would retain for decades: the coexistence of individual adventures with domain, fortress, army, and campaign systems.

They are not necessarily heterogeneous systems glued together later.

They can also be understood as archaeological strata of the different campaign wargaming scales from which the RPG was born.

6. The Blackmoor dungeon¶

At some early point in the campaign, the exploration of the underground space beneath Blackmoor Castle acquired enormous importance.

The exact dating of the first expeditions is still under investigation.

Recently studied material by Blackmoor researchers includes early expedition accounts apparently written by Arneson himself, including:

The Dungeons of Black Moore Castle;
Into the Dungeons of Blackmoor Castle and Back.

The preserved descriptions contain spatial information:

distances;
directions;
rooms;
forks;
stairs;
corridors.

Some of these routes can be reconstructed on surviving maps.

This constitutes an exceptional source because it allows comparing:

GAME DESCRIPTION
           │
           ▼
SPATIAL ROUTE
           │
           ▼
PRESERVED MAP

7. The published Blackmoor maps¶

The First Fantasy Campaign (1977) preserves maps of the early dungeon levels.

These maps present an extremely free geometry.

We observe:

orthogonal corridors;
diagonal corridors;
large oblique axes;
triangular rooms;
trapezoidal rooms;
angled intersections;
corridors that freely cross the grid;
numerous stairs;
connections between different levels.

One annotation is especially important and states:

10' per square

The grid therefore has an explicit metric function.

But the geometry is not bound to follow it.

Walls cross squares.

Corridors maintain their width while cutting the grid diagonally.

Intersections do not need to occur at square centers.

Rooms do not need to correspond to orthogonal cell sets.

OBSERVABLE BEHAVIOR

In these maps: Grid ≠ discretized geometry.

The grid constitutes a metric support on which geometry of greater freedom is represented.

This still does not demonstrate how movement worked during sessions.

8. The discovery of David Megarry's maps¶

Here appears much more important evidence.

David Megarry was one of the original players of Blackmoor and controlled the character H. W. Dumbo.

Megarry preserved maps made during his explorations of the dungeon.

Decades later they were studied and compared with Blackmoor maps by Daniel H. Boggs.

The Oldest Dungeon Maps in D&D History — Hidden in Shadows

Their importance is enormous because they are not simply retrospectively published maps.

They represent the spatial information a player managed to reconstruct during play sessions.

9. How Blackmoor was mapped¶

Megarry later explained to Boggs how the procedure roughly worked.

Players did not have access to Arneson's map.

Arneson verbally provided information like:

distances;
room dimensions;
relative position of doors;
stairs;
exits;
directions.

Players made their own sketches on graph paper.

Later, Megarry transferred those sketches to his notebook.

When geometry proved particularly difficult to communicate, Arneson could draw for them what their characters could observe.

This produces an extraordinarily interesting spatial pipeline:

REFEREE MAP
       │
       ▼
FICTIONAL REALITY
       │
       ▼
ARNESSON DESCRIPTION
distance + direction + dimensions
       │
       ▼
PLAYER PERCEPTION
       │
       ▼
SKETCH DURING PLAY
       │
       ▼
PLAYER MAP

Megarry's map does not necessarily represent the objective dungeon.

It represents the dungeon as the player knows it.

10. The grid as cartographic instrument¶

Megarry's maps provide an extraordinarily strong clue about the meaning of the grid.

They show:

diagonal corridors crossing squares;
oblique walls;
irregular intersections;
rooms not respecting cells;
geometries using graph paper only as a metric background.

Annotations also allow reconstructing approximately a scale of:

1 square ≈ 10 feet

Our initial hypothesis can therefore be refined.

DOCUMENTED FACT

Players used graph paper to map.

Arneson communicated space through distances, dimensions, and spatial relationships.

Maps used a scale.

OBSERVABLE BEHAVIOR

The geometry freely crosses grid units.

STRONG INFERENCE

The grid seems to have functioned mainly as a metric/cartographic instrument, not as an obligatory definition of spatial topology.

NOT DEMONSTRATED

We still don't have enough evidence to know whether certain movement, time, or encounter procedures also used the grid units.

This remains an open question.

11. Diagonals were probably not a special category¶

This allows reconsidering our original question.

From the perspective of a modern digital dungeon crawler we can see:

╲

and immediately think:

DIAGONAL

That implies computational questions:

which Cells does it occupy?
which side is navigable?
how does it collide?
what distance does it cost to traverse?
where can the player rest?
how to render it?
how does it affect the navigation graph?

But nothing demonstrates that Arneson needed such a category.

For him it could have simply been:

a northwest-oriented corridor.

The referee describes:

"the corridor continues X feet toward the northwest".

The player updates their map.

There isn't necessarily any special "diagonal movement" problem.

12. Diagonals as a disorientation mechanism¶

Here we must carefully distinguish observation and intention.

OBSERVABLE BEHAVIOR

Diagonals make manual cartography considerably more difficult.

Modern experiences playing with Blackmoor maps have shown players:

losing orientation;
misinterpreting intersections;
incorrectly closing loops;
believing they visit new spaces when they are returning to known places.

Oblique geometry is therefore objectively capable of producing disorientation.

NOT DEMONSTRATED

We still don't have sufficient evidence to state:

"Arneson deliberately used diagonals to disorient players."

It is perfectly possible.

But there is a simpler alternative explanation.

13. Cartographic hypothesis¶

Arneson proceeded from wargaming.

Campaign maps represent:

paths;
rivers;
coasts;
borders;
cities;
territories.

None of these elements necessarily obey an orthogonal grid.

There are also indications that some early maps related to Blackmoor used European cartography as reference, including Dutch maps.

This allows formulating a hypothesis:

Arneson likely understood the grid as a measurement instrument, not as an ontological structure of the world.

From that perspective, drawing a diagonal corridor is not a mechanical innovation.

It simply means drawing a corridor with the desired orientation.

Disorientation could be a beneficial consequence of that spatial freedom without necessarily being its origin.

14. The referee as spatial processor¶

Here appears probably the most conceptual conclusion of the entire research.

A tabletop RPG can maintain an incomplete spatial representation because there is a human capable of interpreting it.

Arneson could have:

map
+
scale
+
notes
+
world state

The player declared:

"we advance through that corridor."

Arneson could immediately determine:

where they ended;
how much they had advanced;
how much time had passed;
what they could observe;
if they found something;
which part of the map needed to be described.

He didn't need to formally represent all spatial rules before starting the session.

The referee acted as interpreter.

Or, using deliberately modern terminology:

The referee was the Compiler.

He received intention and state, interpreted a partially formal representation and produced a coherent playable reality.

15. From tabletop RPG to CRPG¶

When this type of experience transfers to the computer, the human interpreter disappears.

Then all ambiguities need to become explicit systems.

A computer needs to answer:

where is the player?
what geometry does this space occupy?
can it be traversed?
where does movement end?
what blocks vision?
what surface does the ray hit first?
what region is navigable?
what happens exactly at a corner?

This can be one of the fundamental differences between the spatial representation of tabletop RPGs and computational dungeon crawlers.

In tabletop:

the formal resolution of the system can be inferior to the world resolution because the referee completes the difference.

In a CRPG:

everything that can affect gameplay must eventually become an executable representation.

16. Verticality and connectivity¶

Modern research on Blackmoor maps has also shown that the dungeon had a vertical structure considerably more complex than what can be seen observing published maps alone.

There exist:

stairs;
shafts;
connections between levels;
connections that apparently skip floors;
multiple entries;
alternative routes.

Daniel Boggs has made reconstructions trying to align maps and vertical connections.

Aligning Stairs, Shafts and Elevators in Blackmoor Dungeon

His reconstructions suggest considerable spatial coherence when scales are corrected and different sources are used.

This forces us again to distinguish:

the original dungeon structure

from

the clarity of later published reproductions.

17. A living dungeon¶

Comparisons between:

expedition accounts;
Megarry's maps;
later published maps;

show differences.

Some can be explained through:

unexplored areas;
secret doors;
collapses;
player cartographic errors.

But others seem to suggest that the dungeon evolved during the campaign itself.

This fits perfectly with Arneson's general method:

DESIGN
   ↓
PLAY
   ↓
OBSERVE
   ↓
MODIFY
   ↓
PLAY AGAIN

Therefore, trying to find "the definitive Blackmoor map of 1971" can be as problematic as trying to find "the definitive Blackmoor rules of 1971."

Probably both were evolving.

18. Blackmoor as an emergent system¶

All of the above allows formulating a general interpretation.

Blackmoor seems to have been a system whose complexity gradually arose from the needs of a persistent campaign.

When it was necessary to resolve individual combat:

→ new rules appeared.

When a player wanted to do something not anticipated:

→ the referee adjudicated.

When an idea worked:

→ it could be incorporated into the game.

When the world needed to expand:

→ new places appeared.

When the dungeon needed to grow:

→ geometry was added.

This can explain why it's so difficult to reconstruct Blackmoor as if it were a commercial product.

It still wasn't.

It was a process.

19. Provisional archaeological conclusion¶

Our current description would be:

Blackmoor was an experimental campaign born from wargaming and the Braunstein tradition in which individual characters began to inhabit persistently a world guided by a referee. Its rules, spaces, and procedures evolved through play itself.

And spatially:

The available cartographic evidence indicates that the grid functioned as a scale and representation instrument, but did not force the dungeon geometry to be composed of discrete units.

Therefore:

GRAPH PAPER
     │
     ▼
SCALE / MEASUREMENT
     │
     ▼
CONTINUOUS MAP GEOMETRY
     │
     ▼
REFEREE INTERPRETATION
     │
     ▼
PLAYER EXPERIENCE

And we don't have enough evidence to replace it with:

GRID
  ↓
CELL
  ↓
CELL
  ↓
CELL

This is likely the provisional answer to the question that started this entire excavation.

20. Connection with Lands of Folklore¶

The research began precisely because Lands of Folklore was trying to computationally solve something that Blackmoor could resolve humanely.

LoF preserves a discrete editorial model, but currently can represent:

orthogonal walls;
diagonal geometry;
AIR/MASS regions;
compiled navigation;
internal navigation positions;
separation between editorial intention and Runtime representation.

The pipeline is approximately:

EDITOR
Structural Geometry
         │
         ▼
COMPILER
interprets spatial intention
         │
         ├──► Runtime Geometry
         └──► Compiled Navigation
                     │
                     ▼
                   PLAYER

The conceptual similarity with the tabletop process is quite amusing:

MAP + NOTES
       │
       ▼
    ARNESON
       │
       ├──► spatial reality
       ├──► consequences
       └──► description
                 │
                 ▼
              PLAYER

This doesn't mean that LoF was architecturally derived from Blackmoor.

LoF's architecture was developed independently.

But the historical comparison helps explain why the Compiler results in such a powerful boundary: it performs part of the interpretation work that in a tabletop RPG the referee can do implicitly.

21. A possible future homage¶

The current support for diagonals in LoF opens an interesting possibility.

It doesn't seem possible —nor desirable— to literally reproduce all of Blackmoor's geometry.

Its maps allow angles and compositions that exceed LoF's current structural vocabulary.

But it would be possible one day to make a spatial reinterpretation inspired by its principles:

long diagonal corridors;
oblique axes crossing orthogonal structures;
loops;
strange crossings;
hard-to-map rooms;
vertical routes;
forks;
deliberately uncomfortable orientation.

It wouldn't be necessary to add any feature to LoF just to achieve it.

The interesting test would be precisely:

How much of Blackmoor's spatial language can LoF express using only its current vocabulary?

A future homage to Dave Arneson might lie in the level architecture, not in a textual reference or statue.

22. Questions we leave buried for future excavations¶

There are several issues that still deserve research but don't need to be resolved now:

Original movement: how exactly did Arneson adjudicate movement through corridors during 1971–72?
Mechanical function of the grid: did the squares have any function during exploration beyond scale and cartography?
Diagonals: is there any testimony of Arneson, Megarry, or another original player explaining why they used so much oblique geometry?
Disorientation: was it a deliberate objective or an emergent consequence?
Megarry: asking him directly how they resolved movement and distances through diagonal corridors would be an exceptional primary source.
First expeditions: dating and studying in detail the oldest preserved accounts.
Map evolution: chronologically reconstructing Arneson → Megarry → First Fantasy Campaign.
Ironclads: comparing original rules with early mechanics attributed to Blackmoor.
Chainmail: determining exactly which parts survived after the first sessions.
Tonisborg: using Greg Svenson's dungeon as a potential control to distinguish Arneson's personal habits from a possible "Twin Cities School."
Verticality: representing Blackmoor as a three-dimensional graph of connections instead of a collection of floor plans.
Arneson/Gygax: determining later which components of this experimental system were formalized, modified, or discarded during the creation of D&D.

23. Sources worth preserving in the file¶

The spatial research should especially preserve Hidden in Shadows / Boggswood, because Daniel H. Boggs has worked extensively with Megarry's maps, Arneson's material, and Blackmoor spatial reconstructions.

For the published 1977 maps we have Wayne's Books — First Fantasy Campaign.

The reconstruction of the birth of the dungeon crawl by DMDavid is useful as a secondary source, while Reddit and ENWorld discussions we've found should be preserved mainly as historiographic clues, not as primary evidence.