What we know — and what we still may not automate

Sibilla is not one universal deck or one universal technology.

Auction/catalogue literature supports a May 1827 first-edition/legal-deposit lead, but the primary deposit record is not retained.

The British Museum 1876 catalogue records an early 52-piece La Sybille des Salons set.

The early British Museum set includes a supernumerary explanatory card; an independent c.1830 auction object also records a double-sided instruction card.

The British Museum F.95 catalogue explicitly describes the early series as lacking the characters of true playing cards.

The early explanatory card is catalogued with Alphonse Giroux and Gihaut Frères seller addresses.

A modern researcher reports and transcribes an early instruction-card image attributed to the British Museum; this is a strong lead, not yet retained primary-image evidence.

The recovered secondary transcription says the deck is shuffled before the cut.

The recovered early-instruction lead specifies a cut by the querent with the left hand.

The recovered early-instruction lead uses 35 cards arranged in order on the table.

If the querent-representing card is absent from the 35, the early-instruction lead says to reshuffle the full deck and repeat until it appears.

The early-instruction lead counts by seven beginning immediately after the querent card and interprets every seventh card.

The early-instruction lead divides the remainder into person, house, future and surprise packets.

The early-instruction lead says a reversed card augurs the contrary of its natural-direction announcement.

The exact early-edition card identity used as querent significator is not source-closed; 24/37 may not be back-projected from 1983.

The complete secondary French transcription strongly determines a fixed 35-card circular count by seven; primary-image diplomatic verification is still pending, and the exact remainder passed into the four packets remains open.

A complete 52-card H. Pussey 1848 reprint is documented by Christie’s.

BnF catalogues a complete c.1889 52-card object with blank card and envelope.

BnF preserves a second late-nineteenth-century Salons object suitable for cross-edition collation.

The c.1889 Gallica object has a 109-file public mirror index, but exact bytes and a card/leaf map are not retained yet.

The former arithmetic-only 109-file pairing hypothesis is superseded: the public BnF-derived Commons object is now structurally mapped as 52 alternating front/back card pairs (001-104), blank 105/106, and auxiliary/envelope views 107-109. This structural map does not itself fill unread exact captions.

BnF catalogues a 25-page Art de tirer les cartes avec la sibylle des salons booklet.

A late-nineteenth-century object record physically associates a 52-card Grimaud Salons pack with the Art de tirer booklet.

Playing-card insets are present in later Salons lines but are not universal to the early lineage recorded as lacking true playing-card characters.

The 1983 manual treats upright/reversed orientation as important.

The 1983 manual has the querent wash cards face-down with turning motions to create mixed orientations.

The 1983 manual specifies a left-hand cut.

The 1983 cross method draws 35 cards and repeats if the required significator is absent.

The 1983 manual identifies card 24 for a male consultant and card 37 for a female consultant.

The 1983 manual explicitly removes each seventh card and continues until the consultant card is selected, but the exact machine recount convention is still not uniquely closed.

The 1983 manual deals unselected cards one by one into four cross-positioned piles for existence, home, future and unexpected events.

The 1983 worked example explicitly says the four piles may have unequal sizes depending on discovery/location of the consultant card among the 35.

The documented 1983 ceremony is not proof of the 1827 ceremony, even where structures resemble the early secondary-transcription lead.

Vera Sibilla is stably represented as a 52-card four-suit deck in modern publisher/practitioner sources.

The exact origin/publication year of Vera Sibilla is not source-closed.

The common Solesio/1860 attribution is not accepted as primary-source proven in this pass.

A retained early instruction witness for full card meanings is not acquired.

Historical Vera shuffle/cut/deal rules are not acquired.

Historical Vera layouts are not acquired.

3♦/8♦ titles differ between older and later Morreale/Masenghini-style concordances.

7♥ has a Letterato/Artista title variant across edition lines.

Lottery numbers are treated as a later edition layer, not original universal Sibilla technology.

Printed keyword captions are treated as a later edition layer, not original universal technology.

Straight-card practice is allowed only as a named modern school.

Reversal practice is allowed only as a named modern school unless earlier evidence is acquired.

A current Antica/La Sibilla Italiana 32-card line exists and is separate from Vera 52.

Antica 32-card historical method is not reconstructed from Vera 52.

Juxtaposition/chance/manipulation are historically important in French cartomancy, but not a Vera-specific rule by themselves.

Choosing attractive cards from a fan is not silently substituted for historical sequential procedures.

AI may interpret a fixed draw but may not choose, replace, reorder or reorient cards after the draw.

Physical Deck mode records user-provided physical outcomes only.

The 1983 manual explicitly treats the Cross as one commonly used method and permits practitioners to develop personal procedures; it is not evidence of one universal Salons ceremonial.

The 1983 reading instructions combine each card message, synthesis of the whole packet, encounters between cards, and the positional domain context.

A complete French transcription of the early explanatory card is available through a secondary source that attributes its image to the British Museum; physical existence is independently supported, but primary image provenance is not retained.

The early wording strongly determines a fixed circular row of 35 cards counted by sevens from the card after the querent card until the cycle returns to it.

With 35 cards and step 7, the fixed cycle has five landmarks at ordinal counts 7, 14, 21, 28 and 35; the fifth returns to the querent card.

The early transcription does not instruct removal of each seventh card or a shrinking-deck recount; that mechanic remains specific to later evidence unless independently proven earlier.

The four post-count packets are assigned to person, house, future and surprise.

The exact set of cards meant by the cards that remain before the four-packet cut is not yet source-closed.

The complete 52-card family is institutionally proven, but the normalized 52-title map is not yet an exact retained primary per-card transcription and therefore cannot yet freeze the own-IP identity basis.

A late nineteenth-century object description corroborates the four king identities as Homme de loi/spades, Protecteur/clubs, Militaire/diamonds and Rentier/hearts; this is late-edition evidence, not an early-edition projection.

A dealer-dated c.1860 surviving object and recurring modern Lamperti-1890 reproduction claims do not together prove an exact first publication chronology; Vera origin remains open.

Six J.M. Simon / France Cartes 1979 facsimile sheets directly expose all 52 French labels, numbers 1–52, and playing-card insets; this closes the late facsimile layer, not the primary c.1889 layer.

Direct facsimile collation exposes at least two incompatibilities with the prior normalized registry: no. 10 Partie de Campagne versus Une promenade à la campagne, and no. 43 Chagrin, Larmes versus Chagrins, larmes. They are retained as variants/conflicts rather than silently corrected.

The early 52-card object and F.95 catalogue support an early series without true playing-card characteristics; late suit/rank insets and numbering may not be projected backward without separate primary evidence.

The directly visible late facsimile shows 1 Protecteur / 2 Amie sincère / 3 Flatteur, while a separate secondary history previously reported a different order. Numbering chronology remains open and cannot support final identity freeze.

BnF catalogues the Grimaud booklet as In-24, 25 pages, NAINS-1503; a late physical-object record describes an associated booklet as small in-12, [2]-23-[3] pp. Full text is not acquired and the counting/edition difference remains unresolved.

Despite direct visual transcription of 52/52 late-facsimile cards, the final own-IP historical identity freeze is withheld: primary c.1889 per-card transcription remains 0/52, early numbering is unproven, and edition chronology contains a conflict.

A complete late-nineteenth-century H. Pussey physical deck with an 1890 tax stamp directly shows a profile beginning #1 Amour, #2 Amie sincère, #3 Absence, #4 Attente; this is a distinct edition-bound order.

A complete B.P. Grimaud physical deck around 1890 directly shows #3 Absence, #4 Attente, #8 Chagrin secret, #9 Caquets, and #10 Dépit, corroborating an order incompatible with the 1979 facsimile.

La Sibylle des Salons does not have one universal numbering that can safely be applied across editions; a card number is frozen only within a named edition profile.

Card number and playing-card inset are edition-bound metadata; they cannot be treated as invariant semantic identity across the entire lineage.

The final historical basis of the original 52-card OmenHall Sibilla must name a specific source-edition profile with complete evidence coverage; a mixed “universal” numbering is prohibited.

Similarity between the BnF c.1889 H. Pussey object and the photographed H. Pussey deck in size, square corners, plaid backs, and tax context does not prove the same 52-card order.

Separate complete late-nineteenth-century H. Pussey and B.P. Grimaud physical decks directly agree at least on #3 Absence, #4 Attente, #8 Chagrin secret, #9 Caquets, and #10 Dépit; all five numbers conflict with the 1979 facsimile, providing strong direct evidence for a distinct late-19C numbering profile.

For the BnF/Gallica c.1889 object, 46 exact printed card titles are now directly retained with number and playing-card inset; six exact lower captions remain unread.

For 49 of 52 c.1889 cards, number/scene/inset evidence directly supports semantic identity; cards 1, 22 and 35 remain inference-only until direct capture.

The 109 public BnF-derived files map as 52 front/back pairs in ordinals 001–104, blank 105–106, and three auxiliary views 107–109; this is a navigation structure, not locally retained image bytes.

BnF c.1889 now directly aligns with the Amour-first H. Pussey profile on 19 common exact-title direct positions (with two lexical variants) and with B.P. Grimaud on 4 exact + 1 semantic direct positions; this strongly corroborates one numbering family but does not prove full 52/52 physical-object equivalence.

CCFr records a distinct later B.-P. Grimaud 1967 booklet of 31 pages; it is not a substitute for the undated 25-page BnF NAINS-1503 and is not used to fill unseen text.

For all six unresolved BnF c.1889 cards, high-resolution primary-derived front files and their ordinals/dimensions are individually verified, but the lower captions are not treated as read.

The final named-edition identity freeze for BnF c.1889 formally fails preflight: 52/52 exact direct titles are required, 46/52 are present, and no semantic or cross-edition substitution is allowed.

The BnF catalogue for NAINS-1503 confirms the 25-page booklet and an available reproduction-order route; the reproduction and full text have not yet been acquired into the corpus.

The 25 p. NAINS-1503, [2]-23-[3] pp. associated late object, reported 24-page witnesses, and the distinct 31-page CCFr 1967 item are retained as separate witnesses and are not collapsed without direct collation.

The BnF/Gallica c.1889 named edition is now closed at 52/52 exact printed titles; the final six captions were directly read from exact hash-bound original images.

The final historical identity freeze is issued for the specific BnF/Pussey c.1889 profile: 52 numbers, 52 insets, and 52 exact printed titles. This is not universal Salons numbering and not a historical-method freeze.

OmenHall Salon is an original modern method built on the frozen c.1889 52-card identity layer; it is not represented as the exact NAINS-1503/Grimaud procedure.

All 52 commercial card meanings are newly authored OmenHall expression bound to the c.1889 card identities and are not transcribed historical meanings.