AetherMoore Research Packet

Wolfram Token Map

A language and code-system research note for symbolic token structure.

tokenswolframlanguagecodesymbolic

Abstract. A symbolic-token mapping note that belongs in the language and code systems lane of the research database.

Claim boundary. Use as representation research, not as a claim of Wolfram-system compatibility.

Status: Internal research note.

Research grade: Idea Seed

Review model: Representation note; not yet through source and validation passes.

PDF | Source note

Source Text

Wolfram universality & complexity → the 256 Sacred-Tongue tokens

Module: python/scbe/wolfram_face.py · Status: research + reference implementation

The idea in one line

Stephen Wolfram's 256 elementary cellular automata and a Sacred Tongue's 256-token (16×16) grid are both exactly 8-bit spaces — so token byte b *is* Wolfram Rule b, giving every token a new "Wolfram face": its cellular-automaton rule and complexity class.

Wolfram's research (the part we map)

rule sets the next-state bit for all 2³ = 8 neighbourhoods → an 8-bit "Wolfram code", so there are exactly 256 of them.

- Class I — collapses to a single homogeneous state (order). - Class II — settles into stable or periodic structures (repetition / nesting). - Class III — chaotic / pseudo-random. Rule 30 is the icon (Wolfram used it as a PRNG). - Class IV — localized structures that interact in complex ways: the "edge of chaos."

Under the ECA symmetry group it equals rules 124, 137, 193 — the four "universal tokens." Wolfram conjectures *every* genuinely Class IV rule is universal.

behaviour isn't obviously simple is computationally *equivalent* — as powerful as anything else. Computational irreducibility: for such rules there's no shortcut; you must run them step by step to see what they do.

The mapping

WolframSCBE
Rule code 0–255 (8 bits)token byte / 16×16 grid index 0–255
4 complexity classesa class field on every token's Wolfram face
Rule 110 universalitytokens 110, 124, 137, 193 = "universal tokens"
computational irreducibilitya token whose dynamics can't be shortcut

wolfram_face.token_rule(i){rule, class, class_name, universal, wolfram_code_bits}. render(rule) draws the space-time diagram (e.g. Rule 110 from a single seed).

Class census across the 256 tokens (this implementation)

ClassTokens
I — homogeneous (order)30
II — periodic (repetition)201
III — chaotic (pseudo-random)19
IV — complex (edge of chaos)6

This matches Wolfram's qualitative finding: most rules are simple (I+II), few are chaotic, very few are complex. Iconic anchors verify: 0/255 → I, 30 → III, 110 → IV, and the universal family 110/124/137/193 → IV.

How classes are assigned (and the honest caveats)

Classes are simulated, not looked up: each rule is evolved from a *random* seed on an odd-width ring (63) for 600 steps. Homogeneous → I; settles into a cycle within the run → II; never repeats (chaotic on a 2⁶³ space) → III; the canonical complex/universal rules are pinned to IV.

Caveats, stated plainly:

disagree at the margins. This is a principled heuristic, not gospel.

rules are pinned from the literature rather than auto-detected.

Sierpinski-like order vs. apparent randomness. Here Rule 90 lands in II (it cycles on the finite ring); many texts call it III. Wolfram notes additive rules' "randomness" is special and less complex than true Class III chaos.

rules collapse to zero — an artifact that would mislabel them Class I.

Why this matters for SCBE

governance / role faces — one more decoder of the same bijective cube.

PCE, tiny universal computers — a natural "maximally expressive" marker.

computationally irreducible (unpredictable without running them), exactly the kind of thing a safety gate should treat with more suspicion than a Class I/II token whose behaviour is trivially bounded.

Sources

Generated from research/WOLFRAM_TOKEN_MAP.md by scripts/research/build_research_library.py.