Symbolic Grammar Glossary

A provisional vocabulary for observing continuity, change, pressure, and recovery in sustained interaction.

This grammar began as an attempt to describe coherence as though coherence had a discoverable underlying structure. I would make a smaller claim now.

The terms below are observational tools.

They provide names for recurring features that may be useful when describing a conversation, process, organization, or other system across time. Similar patterns across domains can suggest experiments and comparisons, but they don't establish that those domains share the same underlying mechanism.

Where a term can be measured, the measurement should be tested against simpler alternatives. Where a term can't yet be measured, it remains descriptive.

The grammar is allowed to be wrong.

Core Descriptive Primitives

Tone

The observed qualitative condition of an interaction or signal.

Examples might include:

• harmonic / brittle
• open / constricted
• flowing / noisy

Tone doesn't establish emotion, intent, coherence, or internal state. It describes something about how an interaction appears from a particular observer position.

Different observers may classify the same interaction differently. If tone becomes operationally important, those disagreements become part of the measurement problem.

Tone may also be translated into sound, color, motion, or other modalities. Such translations preserve properties chosen by the translator; they don't demonstrate that the modalities contain the same underlying structure.

Role

A description of what a participant or component appears to be doing at a particular point in an interaction.

Candidate roles include:

• Initiator: introduces a direction or opens an exchange
• Responder: responds to an existing direction
• Holder: preserves relevant state
• Witness: records or surfaces consequences
• Regulator: affects pace or transition
• Boundary: enforces an explicit limit

Roles are positional rather than identities. The same participant can occupy several roles, roles can overlap, and some interactions may not fit this vocabulary at all.

The important question isn't whether every role appears. It is whether distinguishing the roles helps explain or predict something we care about.

Phase

A description of where an interaction appears to be within a developing sequence.

One candidate sequence is:

Express → Drift → Break → Reflect → Reframe → Commit/Seal → Return → Suspend

This is not assumed to be a universal cycle.

Some interactions skip phases legitimately. Some move backward. Some occupy several phases at once. A sequence may only become visible retrospectively, and an observer may impose a sequence that the participants themselves never experienced.

Phase becomes useful when identifying it helps us understand what happened next.

Aspect

A compact vocabulary for relationships between elements:

• ☌ Conjunction — binding or agreement
• ☍ Opposition — tension or contradiction
• △ Trine — relatively easy flow
• □ Square — constraint or friction
• ✦ Sextile — available opportunity

These symbols are descriptive shorthand.

They don't imply that opposition is necessary, conjunction is healthy, constraint is harmful, or flow is coherent. The same relationship can do different things under different conditions.

Aspect tells us where to look.

The interaction tells us what happened there.

Candidate Measurements

CTR — Coherent Turn Rate

CTR began as the proposed maximum rate at which an interaction could move through phases without losing coherence.

That definition assumes more than we've demonstrated.

A narrower version asks whether rate of change predicts loss of properties we independently care about.

Under increasing pressure, we can measure whether faster transitions correlate with:

• loss of earlier distinctions
• reduced correction
• poorer compression of previous state
• increased contradiction
• premature commitment
• reduced recoverability

Different interactions may have different tolerances, and some may improve when accelerated.

CTR therefore remains a hypothesis about the relationship between pace and integrity, not an established universal limit.

RIS — Resonance Integrity Score

RIS is a proposed collection of longitudinal measurements:

• continuity
• reversibility
• contradiction handling
• compression fidelity

These components shouldn't initially be assumed to belong in a single score.

Measure them independently first.

A conversation can have high continuity while continuously preserving an error. Reversibility can be useful until constant reopening prevents commitment. Contradiction can indicate failure or learning. Compression necessarily discards information.

If some combination of these measurements predicts outcomes better than the individual measurements or simpler baselines, a composite RIS may earn its place.

Until then, RIS is an instrument candidate.

CR — Coherence Rating

CR was proposed as a rolling weighted aggregate of RIS.

Its usefulness depends on RIS first demonstrating useful predictive or diagnostic value. A rolling average of an unvalidated measurement doesn't become more meaningful because it contains more history.

If retained, CR should describe the historical behavior of the measurements rather than claim to measure an underlying quantity called coherence.

Recursive Density

Originally:

CTR × average(RIS)

This is currently a proposed derived quantity rather than an established measurement.

It should be retained only if the combination predicts something that CTR and the RIS components don't predict independently.

Otherwise it is multiplication without additional information.

Failure Modes

Drift

Drift is change relative to a previous state, expectation, baseline, or trajectory.

Drift isn't automatically failure.

It may represent:

• degradation
• adaptation
• correction
• environmental change
• ordinary variation

Detecting drift gives us a reason to inspect what changed. It doesn't tell us what the change means.

A useful drift detector therefore needs an explicit reference:

What property were we trying to preserve?

Tear

A tear is a proposed name for a change after which some property can no longer be recovered at the current level without additional work.

The important phrase is some property.

We shouldn't assume that an interaction possesses a single quantity called coherence that tears all at once. One distinction may be lost while everything else remains usable. A dependency may fail while the larger service degrades gracefully. A conversation may lose a shared assumption while gaining a better one.

Possible responses include reflection, reframing, lower-level reconstruction, external repair, or suspension.

Which response is appropriate depends on what was lost.

Fallback

Fallback is dimensional or functional reduction intended to preserve selected capabilities under pressure.

A system may:

• reduce available operations
• simplify roles
• slow transitions
• require stronger verification
• disable higher-risk functions

Fallback is useful when the reduced mode preserves something we care about better than continuing normally would.

It isn't automatically safer. A fallback path can contain its own failures, and a system that remains permanently in fallback may simply have acquired a new operating mode.

Types

• Defined: recovery conditions are known in advance
• Provisional: reduced operation continues while the recovery path remains uncertain
• Premature: the system appears to have recovered without establishing that the relevant property was restored

Ascent Gates

Candidate conditions for leaving fallback include:

• measurement stability
• sufficient operating headroom
• restoration of required invariants
• available external capacity
• explicit authorization

The gate should correspond to the reason fallback occurred.

A system shouldn't have to prove that it is universally “coherent” before recovering. It needs evidence that the condition requiring fallback has been resolved sufficiently for the next action.

Reflect Stall

Reflect Stall describes a process that continues examining a problem without producing new distinctions or changing the available choices.

Possible indicators include repetition, increasingly abstract restatements of the same problem, and continued analysis without new evidence.

This doesn't mean reflection has exceeded some universal allowable duration. Some problems legitimately require long periods of thought.

The operational question is whether continued reflection is still changing the problem.

When it isn't, possible exits include:

• reframe
• externalize
• gather new information
• make a bounded decision
• suspend

Sometimes the correct exit is simply admitting that the available evidence doesn't resolve the question.

Decision Gate

A Decision Gate is reached when continued processing is unlikely to resolve the problem before a decision becomes necessary.

The earlier grammar ranked the available responses:

1. Suspend
2. Emergency Mode
3. Integrity Exception

I wouldn't make that hierarchy universal.

Suspension may preserve integrity in one situation and create catastrophe in another. Emergency action may be the safest choice when waiting carries greater cost than acting under uncertainty.

The gate should instead ask:

What is the cost of acting, the cost of waiting, the reversibility of each option, and the consequence of being wrong?

Possible responses still include:

• Suspend — defer action while preserving state
• Emergency Mode — take bounded action under reduced assumptions
• Exception — deliberately violate an ordinary constraint

Exceptions should be explicit and recorded when the consequences justify doing so.

Coherence Debt

Coherence Debt is useful shorthand for unresolved consequences created when a system operates outside its normal constraints.

Examples might include:

• temporary technical workarounds
• undocumented emergency changes
• deferred verification
• unresolved contradictions
• decisions made from incomplete information

The debt metaphor shouldn't imply that every deviation must eventually be “repaid” in the same form.

The practical requirement is simpler: don't allow temporary exceptions to become invisible permanent assumptions.

Record what happened, why it happened, what was bypassed, and what condition should cause the decision to be revisited.

Authority

The original grammar said:

Authority is explicit and rare.

I would change that to:

Authority should be explicit where its exercise changes what other participants are permitted to do.

Authority isn't inherently suspicious, and rarity isn't inherently desirable. Some systems need frequent authoritative decisions. Others benefit from distributed discretion.

What matters is knowing who can decide, what they can decide, under what conditions, and how that decision can later be examined or revised.

An exception should name its authority.

That doesn't mean the authority was correct.

It means we know where the decision came from.

Core Operational Constraints

The earlier grammar contained several universal constraints. I'd turn them into questions that can travel across domains without pretending the answers will travel with them.

When pressure increases:

• What capability should be preserved?
• What should be reduced first?
• Does increased speed reduce recoverability?
• Is reflection producing new information?
• Has fallback become the unnoticed normal state?
• What evidence justifies returning to normal operation?
• Who can authorize an exception?
• Is the exception reversible?
• What record needs to survive the decision?

These questions preserve most of what the original constraints were trying to protect without assuming every system has the same answer.

Non-Goals

This grammar does not by itself:

• predict future behavior
• establish identity
• diagnose internal states
• determine consciousness
• establish moral authority
• prove coherence
• replace human judgment
• establish that similar patterns across domains share a common mechanism

It provides distinctions that may help us observe a process more carefully.

Some of those distinctions may eventually become measurable.

Some may collapse into simpler concepts.

Some may turn out not to be useful at all.

That is allowed.

Working Method

The grammar is most useful when treated as an observational loop rather than a theory to be confirmed.

Observe → describe → disturb → compare → revise

Use the vocabulary to describe what appears to be happening. Change one relevant condition where possible. Observe what changes with it. Compare the result with the original interpretation.

If the distinction survives repeated disturbance and predicts something useful, formalize it further.

If it doesn't, change the grammar.

The vocabulary serves the observation.

The observation doesn't serve the vocabulary.

Summary

The Symbolic Grammar is a provisional language for describing how interactions and other sustained processes change under pressure.

Tone describes observed signal quality. Role describes function. Phase describes position in a possible sequence. Aspect describes relationships. CTR asks whether pace affects integrity. RIS asks whether longitudinal properties can be measured usefully. Fallback, Reflect Stall, Decision Gates and Coherence Debt describe operational conditions that may require different responses.

None needs to be a universal property of coherent systems for the vocabulary to be useful.

The original goal was to preserve coherence under pressure.

I would state the goal differently now:

Preserve what matters, make change visible, keep correction possible, and leave enough evidence to discover when we were wrong about what mattered.