primary_directions_neo_converse_research - TheDaniel166/moira GitHub Wiki
Primary Directions Research Packet -- Neo-Converse
Purpose
This document records the research path and current admission state of
neo-converse in Moira's primary-directions program.
It records the explicit opt-in branch admitted in September 2026 and the remaining limits on broader historical claims.
It answers four questions:
- what
neo-converseappears to mean in current sources - how it differs from traditional converse
- what evidence is strong enough to trust
- why Moira admitted one explicit law and what remains outside that claim
Current Definition Boundary
The clearest current distinction is:
traditional converse- the significator is carried by the same diurnal rotation to the place of the promissor
neo-converse- a later modern converse doctrine in which the direction is defined against the diurnal rotation
This is the most explicit statement currently in hand from technical software documentation.
Admitted Narrow Signed-Motion Boundary
Moira now also admits signed_primary_motion, but only through the explicit
topocentric_zodiacal_aspect_signed_primary_motion preset. That doctrine is a
source-scoped classifier for Makransky's worked Topocentric zodiacal-aspect
product:
- construct one ordered promissor-to-significator arc
- assign zero latitude to the aspect point and use projected perfection
- wrap the arc to the unique signed shortest displacement in
(-180, 180)degrees - classify positive as direct and negative as converse
- treat numerical coincidence as no event
- fail closed at the directionally ambiguous
180-degree boundary - require explicit non-empty significator and promissor filters rather than implicitly searching a target set that contains the antipodal MC/IC pair
The signed-primary-motion admission does not itself implement
neo-converse. It does not assert motion against the diurnal rotation, does
not transform every method family's arc, and does not materialize a companion
converse arc. Traditional converse continues to exchange
promissor and significator roles. The two products remain separately named and
validated.
Strongest Sources Currently in Hand
1. Martin Gansten on Traditional Converse
The most stable traditional definition available in current evidence is Martin Gansten's formulation:
- direct direction: the promissor is carried by the east-to-west primary motion to the place of the significator
- converse direction: the significator is carried by the same motion to the place of the promissor
Source:
This gives Moira a clear traditional baseline.
2. Delphic Oracle Release Notes
The most explicit neo-converse statement currently found is from Delphic
Oracle's release notes:
- neo-converse directions are "defined as directions against the diurnal rotation"
- traditional converse directions are those where the significator is moved with the diurnal rotation
- the option was added to match Morinus software values
Source:
This is strong as a software-technical definition, but it is still not the same as a fully published governing law.
3. Delphic Oracle Product Documentation
Delphic Oracle also states:
- it distinguishes traditional converse from neo-converse
- it follows conventions attributed to Martin Gansten's course
- it uses
neo-converseas a separate direction mode in the UI
Sources:
These sources support the claim that neo-converse is a live modern doctrine
label, not a hallucinated term.
4. AstroApp as Software-Landscape Confirmation
AstroApp confirms that:
- both
traditional converseandneo-converseare exposed as separate primary-direction options in current software
Sources:
This is useful landscape evidence, but not a governing formula.
What Seems Stable
The following now look stable enough to state:
neo-converseis a real modern primary-directions label- it is not identical with traditional converse
- the core conceptual distinction is:
- traditional converse: same diurnal rotation, moving the significator
- neo-converse: direction against the diurnal rotation
- some software treats
neo-converseas the mode needed to match Morinus-style modern values
Admission Evidence and Remaining Limits
The admitted runtime branch now has:
- an explicit circle-complement law
- a distinct typed policy identity
- cross-method invariants for symmetric and asymmetric geometries
- Python/native parity coverage
The remaining research need is stronger external numerical evidence across historical schools and asymmetric method families. That limitation constrains claims of historical universality; it does not make the runtime policy ambiguous.
Risk Assessment
Source Quality
- current standing:
medium for historical universality
Why:
- the concept is clearly documented in active software
- the traditional baseline is clear
- the admitted computational law is explicit, while its broader historical universality remains less strongly sourced
Mathematical Recoverability
- current standing:
implemented narrow law
Why:
- the conceptual difference and computational transformation are explicit
- additional historical variants, if any, remain outside the admitted law
Implementation Risk
- current standing:
low for runtime ambiguity; medium for historical claims
Why:
- the explicit opt-in identity prevents collision with traditional converse
- the remaining risk is overstating one modern convention as historically universal
Moira Policy & Admission (September 2026)
Moira now admits neo_converse under explicit, opt-in policy control:
traditional converse(traditional_converse) remains the default historical doctrine throughout Moira (significator carried to promissor in diurnal rotation; computed via Morin's role-exchange theorem $\text{converse}(S \to P) = \text{direct}(P \to S)$).neo_converse(neo_converse) is formally admitted as a first-class converse motion doctrine:- Motion Law: bodies are directed against the diurnal rotation (West-to-East) while the significator remains stationary as receiver.
- Geometric Derivation: $\Delta_{\text{neo}} = (360^\circ - \Delta_{\text{dir}}) \pmod{360^\circ}$.
- Equivalence & Divergence: On symmetric equatorial/meridian systems (e.g. Meridian), $\Delta_{\text{neo}} = \Delta_{\text{conv}}$ identically. On asymmetric oblique semi-arc systems (Placidus, Campanus, Topocentric), $\Delta_{\text{neo}} \neq \Delta_{\text{conv}}$ because each body moves in its own oblique semi-arc under role-exchange, whereas neo-converse directs the promissor in its own semi-arc along the counter-diurnal circle complement.
- Exposures: Admitted in
PrimaryDirectionConverseDoctrine.NEO_CONVERSE,PrimaryDirectionsPresetconfiguration viaconverse_doctrine="neo_converse", and native C++ solvers with mode'N'.
signed_primary_motionremains confined to the named Makransky Topocentric zodiacal-aspect preset.
Implementation & Verification
The doctrine was implemented in September 2026 and verified via:
- Mathematical circle complement proof: $(\Delta_{\text{dir}} + \Delta_{\text{neo}}) \pmod{360^\circ} \equiv 0^\circ$.
- Asymmetry proof against traditional converse on Placidus mundane directions: $\Delta_{\text{neo}} \neq \Delta_{\text{conv}}$.
- Exact numerical identity with traditional converse on Meridian right ascensional directions: $|\Delta_{\text{neo}} - \Delta_{\text{conv}}| < 10^{-12 \circ}$.
- Validated by unit test suite
tests/unit/test_primary_directions_neo_converse.pyand native parity suitetests/unit/test_native_primary_directions_parity.py.
Present Declaration
neo_converse is fully formalized, admitted, implemented, and verified across Python and native C++ substrates with complete cross-method rigor.