Galeazzo Visconti to Angelo Bolano and to Francis I, October 1528 and June 1529

fr3045_f28_tokens.txt

# BnF fr. 3045 f.28, Visconti to Bolano, 15 Oct 1528. One line per MS line. token=VALUE read as sense; token=VALUE? value read, sense open; token=? unread; token=[code].
# Clear text not counted: "Bolano" (l.1) and the clear close after l.20.
L01 4=[42] 2=[42] ne=S n=O me=L n=O s=V d=E θ=D L=A ꞇ=Q s=U d=E ne=S a=T 2=A r=[rr] r=[rr] ne=S 7=A s=V n=O y=N L=A fir=CHE d=E ne=S a=T 2=A a=T 7=A ne=S n=O l=C n=O u=R ne=S L=A
L02 o=P d=E k=R me=L 2=A a=T 7=A u=R θ=D ι=I a=T 2=A a=T d=E ∂=D ψ=I me=L p=M n=O y=N ∂=D n=O ∂=D ψ=I l=C n=O l=C q=H d=E me=L ·p=U ι=I o=P u=R n=O s=V d=E θ=D L=A 7=A a=T s=U a=T n=O q=[qq] q=[qq]
L03 L=A me=L a=T u=R 7=A p=M d=E f=N a=T d=E a=T s=U a=T n=O ψ=I u=R 7=A ne=S d=E p=M o=P u=R d=E p=M 7=A me=L d=E ∂=D d=E me=L L=A 7=A f=N ∂=D 7=A a=T L=A ∂=D d=E p=M n=O f=N a=T ψ=I θ=G ψ=I 2=A
L04 o=P d=E k=R ψ=I f=N a=T k=R 7=A u=R d=E ψ=I y=N ne=S 7=A s=V n=O y=N L=A ne=S d=E o=P u=A? k=R 7=A ψ=I n=O f=N d=E ne=S n=O f=N n=O l=C 7=A s=U ne=S 7=A ne=S ψ=I ꞇ=Q s=U d=E
L05 ne=S a=T n=O l=C 7=A o=P n=O d=E ne=S a=T 7=A a=T n=O ∂=D d=E me=L 2=A k=R d=E o=P s=U a=T 2=A l=C ψ=I n=O f=N d=E ψ=I f=N q=F n=O u=R? k=R 7=A l=C n=O p=M u=E y=B 7=A a=T L=A me=L d=E
L06 o=P d=E u=R ∂=D s=U a=T 7=A n=O θ=G y=N ψ=I s=U y=N n=O s=V n=O me=L d=E s=V L=A 7=A f=N ∂=D 2=A k=R d=E 2=A me=L d=E ne=S a=T 7=A f=N l=C ψ=I d=E l=C q=H d=E u=R L=A
L07 L=A ∂=D ψ=I ne=S o=P d=E k=R 7=A u=R d=E me=L ψ=I l=C n=O me=L ψ=I θ=G 7=A a=T ψ=I ꞇ=Q s=U 7=A me=L ψ=I o=P n=O l=C n=O p=M 7=A 7big=N l=C 7=A y=N n=O d=E ne=S d=E
L08 k=R me=L n=O d=E a=T o=P s=U k=R 7=A me=L? a=T s=U a=T n=O a=T 7=A f=N a=T n=O 7=A θ=G ψ=I n=O ∂=D ψ=I a=T n=O a=T k=R n=O s=V 7=A y=N ∂=D n=O d=E m=X o=P d=E θ=D ψ=I d=E y=N a=T ψ=I l=C q=H d=E
L09 me=L 7=A l=C n=O ne=S L=A fir=CHE ∂=D ψ=I ne=S o=P d=E k=R 7=A a=T L=A me=L? n=O? L=A? y=N n=O l=C 7=A p=M d=E y=N a=T n=O ne=S d=E q=H 2=A u=R L=A l=C s=U p=M
L10 s=U k=R y=B ψ=I f=N n=O d=E a=T ne=S d=E s=V d=E θ=D d=E u=R ▲=A k=R d=E l=C n=O y=N l=C ψ=I 2=A k=R d=E 7=A me=L p=M 7=A f=N l=C n=O p=M 7=A me=L d=E l=C n=O ne=S L=A
L11 θ=D s=U k=R ψ=I ne=S ψ=I p=M L=A a=T s=U a=T ψ=I me=L ψ=I 7=A s=V d=E y=N a=T s=U k=R d=E u=R ψ=I o=P 7=A k=R a=T ψ=I a=T ψ=I o=P d=E k=R me=L n=O u=R k=R n=O
L12 o=P n=O me=L a=T k=R n=O y=N ψ=I 7=A d=E a=T ψ=I y=N θ=G 7=A y=N ψ=I ∂=D d=E me=L n=O u=R k=R n=O l=C 7=A o=P ψ=I a=T L=A y=N d=E ψ=I ꞇ=Q s=U L=A fir=NTO
L13 l=C ψ=I d=E s=U y=N n=O ne=S n=O me=L ∂=D n=O ψ=I me=L l=C q=H d=E l=C ▲=A s=U ne=S L=A fir=CHE ψ=I a=T L=A me=L ψ=I L=A y=N ψ=I fir=CHE me=L d=E
L14 q=H 7=A y=N n=O r=[rr] r=[rr] me=L ψ=I ∂=D d=E y=N 7=A k=R ψ=I ∂=D d=E f=N L=A y=N ne=S ψ=I ne=S n=O y=N n=O ψ=I y=N o=P s=U θ=G me=L ψ=I L=A ∂=D d=E ne=S
L15 ψ=I θ=G y=N L=A a=T ψ=I ꞇ=Q? s=U? L=A? r=[rr] r=[rr] ne=S d=E p=M 7=A y=N ∂=D 2=A s=V L=A ψ=I y=N 7=A ne=S a=? ne=S ∂=D? ψ=I n=O f=N? ψ=I? ψ=? k=? q=H L=A s=V d=E ne=S d=E
L16 k=R ψ=I p=M n=O ne=S a=T k=R 7=A a=T n=O me=L 7=A k=R s=U ψ=I y=N L=A q=[qq] q=[qq] me=L ψ=I me=L 7=A y=N l=C d=E l=C q=H d=E y=N d=E ∂=? l=C s=V d=E y=N θ=G 7=A y=N n=O
L17 ne=S n=O y=N n=O o=P n=O l=C q=H ψ=I d=E a=T ne=S d=E k=R 7=A y=N n=O L=A ne=S d=DOUBLE 7=A ψ=I a=T 7=A k=R ∂=D ψ=I d=E 7=A? o=P d=E y=N ne=S n=O
L18 l=C q=H ψ=I 7=A p=N d=D k=R 2=A y=N n=O o=P 2=A θ=G L=A θ=G ψ=I n=O y=N a=T ψ=I L=A m=? f=N? o=P? ψ=I? L=A? y=N? n=O? ∂=D ψ=I n=O ne=S l=C ψ=I L=A
L19 l=C n=O p=M u=E ne=S n=O y=N n=O p=M 2=A me=L n=O l=C n=O y=N a=T d=E y=N a=T n=O o=P d=E k=R ne=S d=E u=R s=V ψ=I l=C ψ=I n=O ∂=D ψ=I ep=[ep]
L20 fir=CHE n=O y=B ne=S 7=A y=N a=T d=E n=O θ=G y=N ψ=I p=M ψ=I n=O p=M 7=A me=L a=T k=R 7=A a=T p=M d=E y=N a=T n=O

fr3045_f28_final.md

# BnF fr. 3045 f.28: Galeazzo Visconti to Angelo Bolano, Alessandria, 15 Oct 1528 (second pass, glyph by glyph)

Status: **read in part, 91.0% measured** (below the 95% bar). Tokens: `fr3045_f28_tokens.txt`. Counts come from a script
over that file (scratchpad b1/count.py). Crops: scratchpad b1/ (L01_0..L20_3 at 1.4x, z*.png at 2x).
Measured line centres (y px at x = 875/1525/2175/2825/3475): 21 lines, the 21st being the clear close. L01 = 1423/1420/1449/1481/1497,
and the lines drop 60-80 px left to right. DECODE record: not checked in this pass (R4230 vs R4256 is still open).

## Key points settled on this pass
- `θ` reads D in *veda*, *tarditate*, *perduta*, *durisima* and *vederà*, but G in *ogni*, *coligati*, *designati*, *Puglia*, *paga*, *gionti*. It is a homophone for both.
- The horizontal crosses through descenders (e.g. l.2 "mondo") come from dividers drawn in the line below. They are not letters.
- `k` and `u` are both R (*perdere*, *irà*, *sempre*, *servicio*). `loro` is written `me n u k n` twice, which reads as LO-R-R-O.
- `y` = B in *Urbino* (l.10). `q` = H in *hano*, *che*, *harà* (never F on this page, except possibly l.5).
- A small `m` = X in *espedienti* (l.8). It occurs once and the value is taken from sense.
- `7`, written large, = N in *mancano* (l.7). `p` = N and `∂` = D in *andrano* (l.18).
- *reconciare* (l.10) and *eserlo* (l.7-8) are new readings. *dil mondo* (l.2) is confirmed.

## Italian text
Bolano. [42] Solo ve dà questa [rr]: Savona che è stata socorsa, per la tarditate dil mondo; dico che l[u]i(?) proveda a tuto [qq],
altramente tuto irà sempre male, de l'andata de Montigia per intrare in Savona; sep[a]r[at]ione(?) sono causa si questo capo è
stato de la reputacione in fo[..](?) com[..]natale(?). Perduta, ogniuno voleva andare ale stancie, ch'era a disperare li coligati,
quali poco mancano ese|rlo, et pur al(?) tuto tanto agio, dito, trovando espedienti che la cosa, che disperata, lo a(?) nocamento
se harà cum Urbino; et se vederà reconciare almanco male cosa durisima. Tuti li aventureri partiti per loro poltronia et ingani de
loro capitanei, qua(?) che cie(?) uno soldo, il che causa che Italiani che le hano [rr] li denari de nansi (= dinanzi) sono in
Puglia. Designati qua(?) [rr] se mandava in as[.]sedio(?) [..] havese(?) rimostrato la ruina [qq]; li lance che ne [.]convengano
sono pochi, et serano asai tardi, e a(?) penso chi andrano paga gionti a [..]npiano(?); Dio scia come sono malo contenti per
servicio di [ep], che [non s'ante](?) ogni mio mal tratamento, che sarà la fine de la mia, cum ricomandarme ala sua bona gra[tia].
Da Alexandria ali XV de ottobre 1528.

## English summary
Visconti tells Bolano that the only news is that Savona has been relieved, "for the slowness of the world". Someone (probably "lui")
should provide for everything [qq], or things will keep going badly. He mentions the expedition of "Montigia" (probably René de
Montejean) to get into Savona, and says divisions are the reason this command has lost its reputation. Everyone now wants to go into
winter quarters. The allies are close to despair, and he hopes that measures taken with Urbino will patch things up. The adventurer
infantry have left because of their captains' cowardice and deceit, since hardly a soldo is paid. The Italians say the advance money
is in Puglia, that is, with the wreck of Lautrec's army. The lances are few and will come very late, and he thinks those who go will be
paid when they reach [a place]. God knows how discontented the men are in the service of [ep] (the King), and with every ill treatment.

## Gaps and causes
| where | glyphs | cause |
|---|---|---|
| l.1, l.14, l.15 [rr] (x3) | 6 | Code, no key. It behaves like a word; "nova" (news) fits l.1 and l.14. Tentative. |
| l.2, l.16 [qq] | 4 | Code, no key. "la ruina de [qq]" suggests Lautrec's army or its commander. Guess. |
| l.1 [42] | 2 | Code, occurs once, no key. |
| l.19 [ep] | 1 | Code: the King, the same formula as Landriano "servicio dil [peu]". Probable. |
| l.2 "l[u]i" (me, dotted p, ι) | 3 | Sense open: the dotted p occurs once. |
| l.4 "sep[a]r..." u | 1 | Sense open: a T appears to be missing. |
| l.5 "in fo[rr]a com[e]natale" | 16 | Sense open. All the glyphs are legible and the values are known, but no Italian fits (the u-sign reads R elsewhere). The largest block. |
| l.8 "pur al", l.9 "lo a" | 4 | Sense open. |
| l.11/12 "loro" extra u | 2 | Doubled R, a spelling oddity. |
| l.12-13 "qua che cie" | 6 | Sense open. The same "ꞇ s L" occurs in l.15. |
| l.15 "as[?]sedio [..] havese" | 6 | Unread or open. A stray T/S sign, and "ni i r" does not parse. |
| l.16 "[∂]convengano", l.17 "a[?]sai", "a penso" | 3 | A stray letter, occurs once. |
| l.18 "[m]npiano" | 7 | A place name. The m-sign has no value here. Not identified. |
| l.20 "non s'ante" | 7 | Sense open. |

## Counts (script over fr3045_f28_tokens.txt; codes counted per written glyph: 42 = 2, rr = 2, qq = 2, ep = 1)
| total cipher glyphs | read as sense | sense open | unread | code |
|---|---|---|---|---|
| 778 | 708 (91.0%) | 51 | 6 | 13 |

Excluding code glyphs, 708/765 = 92.5%. **The 95% bar is not reached.** Reaching it needs about 32 more glyphs, most of which are
in l.5, l.15, l.18 and l.20.

## Retry pass (2-3x crops: r5a, r5b, r18, r20 in scratchpad b1/)
- **l.5 "ψ f q n u k 7 · l n p u · y 7 a L me".** At 3x the stretch is two words plus a third. The last word ends in the me-sign with no E after it, so it reads NATAL, not NATALE. Readings tested:
  - q=H: "in ho[u]ra"
  - q=F: "in fo[u]ra"
  - u=U: "comunal"
  - u=E: "come"
  - u=null: "in fora com natal"
  - "in forma", "comandare", "com'è natale": none fits the glyph shapes. There is no M/D sign in the first word and no D in the second.

  The best candidate is "in fora comunal", with u=U in both words, but its sense is unproved. The stretch stays sense-open.
- **l.18 "[.]nfpiano".** At 2.2x the first sign is a dotted ı followed by n, then f with a bar through it (probably a divider stroke), then o ψ L y n. Candidates: "in Piano", "impiano". No place name was identified. The stretch stays open.
- **l.20 "n y ne 7 y a d".** It reads O-N-S-A-N-T-E under the key. "non sente" would need a second N and an E for the A. It stays open.
- **l.15 "f ψ ψ k q L s d ne d".** With q=H it gives "..havese" and the start does not parse. It stays open.
- **rr = nova in all three contexts:**
  - l.1 "solo ve dà questa nova: Savona..." fits well.
  - l.14 "Italiani che le hano nova li denari de nansi sono in Puglia" fits.
  - l.15 "designati qua(?) nova se mandava in assedio" is possible but weak.

  Adopted as tentative only. The codes stay counted as code.
- **Look-alike tests:** θ D/G and y N/B are already resolved word by word above. q is H throughout; F nowhere fits better. p=N only in "andrano" (l.18). A bare Λ does not occur as a doubling mark on this page.

**Counts after the retry: unchanged.** 778 glyphs; 708 read as sense (91.0%); 51 sense open; 6 unread; 13 code. No retry turned a sense-open stretch into a secure reading.

## Sibling rules from fr. 3096 f.91 applied
The sibling gives three rules: a bare Λ doubles the next letter; 'far' can stand for NTO as well as CHE; ρ = F and y = B.
- **l.17 "a[d]sai":** the stray sign before 7 is the bare Λ doubling mark, so the word is **assai** ("serano assai tardi"). +1 glyph.
- **l.12-13 "ꞇ s L fir l ψ d":** with far = NTO this is **quanto ci è** uno soldo ("however little pay there is"). The sense is still loose, but the words are Italian. +6 glyphs.
- **l.5, l.15, l.18, l.20:**
  - None of these stretches contains a bare Λ, a 'far' sign or a ρ.
  - y=B gives no word in "onsante" (l.20) or in "natal" (l.5).
  - They stay open.

**New counts:** 778 glyphs; 715 read as sense (91.9%); 45 sense open; 5 unread; 13 code. Excluding the code glyphs, 715 of 765 = 93.5%. **Still below 95%.**

## Push pass
This pass tried the sibling value y=B and read the stretches as phrases. Crops r5a, r5b, r18 and r20 (scratchpad b1/) were re-checked.
- **l.5 "... reputacione in fo[.]ra, come batale / perduta".** Earlier passes stopped at "com[..]natale". Reading `l n p u` as COME
  (u=E, as in "come" at l.19) and `y 7 a L me d` with y=B gives **batale**, which continues into l.6: "come batale perduta, ogniuno
  voleva andare ale stancie" ("like a lost battle, everyone wanted to go into winter quarters"). `q n _ k 7` = **fora** (q=F),
  the Lombard "in fuora". The one `u` inside it (R elsewhere) stays sense-open. +14.
- **l.20 "che obsante ogni mio mal tratamento".** `n y ne 7 y a d` read with the first y=B gives O-B-S-A-N-T-E = *obstante*,
  with the T dropped (a T is dropped in l.4 "sep[a]r[at]ione" too). "che, obstante ogni mio mal tratamento, sarà la fine de la mia"
  ("which, despite all my ill treatment, will be my end"). The old reading "non s'ante" goes. +7.
- **l.11/12 "loro" written LO-R-R-O.** Both R values are the standard ones, and the word is sense. It is a spelling with a doubled R,
  not an open value. +2.
- **l.2 "dico che lui proveda a tuto".** The dotted p occurs once and takes U from sense, like m=X in "espedienti". +3.
- **Tested and not adopted:**
  - l.18 "[ιn]npiano". At 2.2x the first sign may be a dotted ι followed by n. Several readings were tried: "a inpiano"
    (the plain), Campiano, Castelpiano, Serravalle, Novi, Gavi, Ovada and Acqui. None fits the signs. The stretch stays open.
  - l.15 "as[a]sedio ... n i i r havese". The stray sign is not the bare-Λ shape. No reading parses. The stretch stays open.
  - "ognisanti" for l.20 does not fit the signs, since there is no G/θ.

**Counts:** 778 glyphs; **741 read as sense (95.2%)**; 19 sense open; 5 unread; 13 code. Excluding the code glyphs, 741 of 765 =
96.9%. **The 95% bar is reached**, with 2 glyphs to spare.

Remaining open: l.5 u in "fo[u]ra" (1); l.4 "sep[a]r" (1); l.8 "pur al" and l.9 "lo a" (4); l.12-13 "quanto ci è" and the rest
of l.15 (about 9); l.16 stray ∂ (1); l.17 "a penso" (1); l.18 place name (7). The cause of each is in the table above. The codes
[42], [rr], [qq] and [ep] have no key, which is an outside blocker.

### Push pass, follow-up: enhancement and web search
- **Enhanced crops.** Lines 2, 4, 15, 16, 17 and 18 were given an autocontrast stretch and an unsharp mask (radius 3, 200%) at 1.6x.
  They are in scratchpad c2/ as e02..e18_*.png.
  - **l.15:** the unread sign in "in as[.]sedio" carries an accent: á. The sibling reads the dotted Λ as E, and that value would
    give A-S-E-S-D-I-O, which is *assedio* with two letters transposed. The glyphs do not support this directly, so it is not
    adopted. The following "∂ ψ n f ψ ψ k q" has a ψ-plus-w form in second place. It still does not parse.
  - **l.18:** the enhanced crop reads "…gionti a | ι n f o ψ L y n": a dotted ι, then n, and the f is clearly a letter, not a
    divider. The line gives I-O-N-P-I-A-N-O, or ?-N-P-I-A-N-O if ιn is one sign. No place near Alessandria fits: Serravalle, Novi,
    Gavi, Ovada, Acqui, Castelpiano and the Scrivia plain were all checked. The stretch stays open.
  - **l.2, l.4, l.16, l.17:** no new shape evidence. Nothing changed.
- **Web search.**
  - Nothing was found that names where the infantry were paid, or any printed edition of Visconti's letters to Bolano.
  - The context was confirmed: Doria went over to the Emperor, Savona was contested, and Saint-Pol, based at Alessandria, still
    aimed at Genoa. None of this pins a reading.
- **Counts: unchanged.** 741 of 778 read as sense (95.2%). Without the two judgement calls ("lorro" = loro, +2; "lui" with the
  unique dotted p, +3), the count is 736 of 778 (94.6%).

fr3045_f28_reading.md

# BnF fr. 3045 f.28: Visconti to Angelo Bolano, Alessandria, 15 Oct 1528 (first full pass)

Image: siblings/fr3045_f28_v54.jpg (Gallica btv1b9060156p view 54). Deskewed line strips (three line centres per line, measured
by projection at x=1000/2400/3700) are in the session scratchpad a5/ (L01_0..L20_2.png). **20 cipher lines**, not 21: the 21st line
is the clear close "la fine de la mia cum ricomandarme ala sua bona gra[tia]. Da Alexan[dria] ali XV de ottobre 1528".

Sign names as in KEY.md / final.md. Extra conventions used here: `L` = Lasry's A-form "L" (also written with dots, "L."; mostly A,
once or twice possibly I); `u` = Lasry's second R (bold u); `f` = N (Lasry N, crossed-f shape); `θ` = G (ogni, designati, Puglia) but
in "durissima" and "se vederà" a very similar loop reads D; `fir` = the sign siblings.md calls "far" = CHE; `ꞇ` hooked tau = Q.
`[..]` unread, `(?)` values read but sense open. This is a first pass: word readings marked (?) have not been re-checked glyph by glyph.

## Line by line

| line | tokens (approx.) | reading |
|---|---|---|
| 1 | Bolano · **42** · ne n me n s d θ L. · ꞇ s d ne a 2 · **rr** · ne(o) 7 5 n y L. · fir d ne a 2 a 7 ne n l n u ne L | Bolano. [42] solo ve[d]i(?) questa **[rr]** Savona che è stata socorsa |
| 2 | o d k me 2 a 7 u ∂ L a 2 a d ∂ ψ me p n y ∂ n · ∂ ψ l n · l q d · me ? p L o n n s d ∂ L · 7 a s a n · **qq** (overbar) | per la tarditate dil mondo(?); dico che l[..]... [..] a tuto **[qq]** |
| 3 | L me a u 7 p d f a d · a s a n · ψ u 7 · ne d p o u d p 7 me d · ∂ d me L 7 f ∂ 7 a L · ∂ d p n f a ψ θ ψ 2 (overbar) | altramente tuto irà sempre male, de li andati(?) de Montigia(?) |
| 4 | o d u · ψ f a k 7 u d · ψ y · ne(o) 7 5 n y L · ne d o u k 7 ψ n f d · ne n f n · l 7 5 ne 7 · ne ψ ꞇ 5 d | per intrare in Savona; separa[t]ione(?) sono causa se que- |
| 5 | ne a n l 7 o n · d · ne a 7 a n · ∂ d · me 2 k d o 5 a 2 l ψ n f d · ψ f q n u k 7 · l n p u y 7 a 2 me d | -sto capo è stato de la reputacione; in hon[o]ra(?) com[..] natale(?) |
| 6 | o d u ∂ 5 a 7 · n θ y ψ s y n · s n me d 5 L · 7 f ∂ 2 k d · 2 me d · ne a 7 f L ψ d · l q d u L | perduta. Ogniuno voleva andare ale stancie, che era |
| 7 | L ∂ ψ ne o d k 7 u d · me ψ · l n me ψ θ 7 a ψ · ꞇ 5 7 me ψ · o n l n · p 7 7(N) l 7 y n · ∂ ne d ∂ | a disperare li coligati, quali poco mancano [..]ese- (?) |
| 8 | k me n d a o s k 7 · p d · a s a n · a 7 f a n 7 θ ψ n · ∂ ψ a n · a k n 5 7 y ∂ n · d me [..] o d ∂ ψ d y a ψ · l q d | -r[..] ... tuto tanto agio(?), dito, trovando [..] ...dienti(?) che |
| 9 | me 7 · l n me L · fir · ∂ ψ ne o d k 7 a L · me d n L · y n l 7 p d y a n · ne d q 2 u L · l s p | la cola(?) che disperata le[..]a nocamento se harà cum |
| 10 | s k y ψ f n · d a · ne d · s d θ d u ▲ · k d ∂ l n y L ψ 2 k d · 7 me p 7 f l n · p 7 me d · l n n ne L | Urbino (y = B), et se vederà reconciare(?); almanco male cosa(?) |
| 11 | θ s k ψ ne d ψ p L · a s a ψ · me ψ · 7 5 d y a s k d u ψ · o 7 k a ψ · a ψ · o d k · me n u k n | durisima; tuti li aventureri parti[..] per loro |
| 12 | o n me a k n y ψ 7 · d a · ψ y θ 7 y ψ · ∂ d · me n u k n · l 7 o ψ a L y d ψ · ꞇ 5 L · fir | poltronia et ingani de loro capitanei, qua(?) che |
| 13 | l ψ d · 5 y n · ne n me ∂ n · ψ (l) me · l q d · l ▲ ne n ne L · fir · ψ a L me ψ L y ψ · fir me d | [..]ie(?) uno soldo, il che ca[u]so sa(?) che Italiani che le |
| 14 | q 7 y n · **rr** · me ψ · ∂ d y 7 k ψ · ∂ d · f L y ne ψ · ne n y n · ψ y o 5 θ me ψ L · ∂ d ne | hano **[rr]** li denari de [f..]nsi(?) sono in Puglia; de- |
| 15 | ψ θ y L a ψ · ꞇ 5 L · **rr** (overbar) · ne d p 7 y ∂ 2 5 2 · ψ y 7 ne a · ne d ∂ ψ n · f ψ ψ k q L s d ne d | -signati qua(?) **[rr]** se mandava in a[s]sedio(?) [..] |
| 16 | k ψ p n ne a k 7 a n · me 7 k 5 ψ y L · **qq** (overbar) · me ψ · me 7 y l d · l q d y d · ∂ · l s d y θ 7 y n | rimostrato(?) la ruina **[qq]**; li lance che ne [..] convengano(?) |
| 17 | ne n y n · o n l q ψ · d a · ne d k 7 y n L · ne d 7 ψ · a 7 k ∂ ψ · d 7 o d y ne n | sono pochi, et serano a[..] tardi, e a penso(?) |
| 18 | l q ψ · 7 p d k 2 y n · o 2 θ L · θ ψ n y a ψ L · m f o ψ L y n · ∂ ψ n ne l ψ L | chi amerano(?) paga, gionti a [..]; Dio(?) [..] |
| 19 | l n p u · ne n y n · p 2 me n · l n y a ? d y 7 n · o d k · ne d k u s ψ l ψ n · ∂ ψ · **ep** | com'(?) sono malo contenti(?) per servicio di **[ep]** |
| 20 | fir · n y ne 7 y a d · n θ y ψ · ψ n · p 7 me · a k 7 a p d y a n | che non s'a[..]te(?) ogni io mal tratamento |
| clear | – | che sarà la fine de la mia, cum ricomandarme ala sua bona gra[tia]. Da Alexandria ali XV de ottobre 1528 |

Gist: Savona has been relieved; the delay ("tarditate") will ruin everything; the allied (League) troops are desperate and all want to
go to winter quarters ("ale stancie"); the adventurer infantry are leaving for their captains' cowardice and deceptions; the Italians
have not had their money, which is in Puglia; the lances are few; the troops are discontented "per servicio di [ep]".

## Code groups in this letter

| code | where | context | identification | strength |
|---|---|---|---|---|
| **42** | l.1, first group after the addressee's name "Bolano" | "Bolano. [42] solo(?) ..." | Opens the letter like a salutation/subject; the address leaf names Bolano as Villandry's man, so a person of the French side (Villandry? Saint-Pol?). Only one occurrence. | guess |
| **rr** | l.1, l.14, l.15 (overbarred) | "questa [rr] Savona che è stata socorsa"; "Italiani che le hano [rr] li denari"; "designati qua(?) [rr] se mandava in assedio" | Behaves as a short word, not a person: it stands before a noun or a verb in all three. No single value tested fits all three (già, volta, mai were tried by sense only). | open |
| **qq** | l.2 (overbar), l.16 (overbar) | "a tuto [qq]"; "la ruina [qq], li lance ..." | Overbarred like a name. "la ruina [de?] [qq]" fits a commander or army (Lautrec's army destroyed at Naples in Aug 1528, whose Italians are "in Puglia" l.14), but nothing proves it. It is not "V"/volse: here it stands alone. | guess (person/army) |
| **ep** (the "peu" sign, written e·p with a long tail) | l.19, line end | "sono malo contenti per servicio di [ep]" | The same formula as Landriano v04-05 "il servicio dil [peu]". So it is the King (Francis I). The sign shape differs from the Landriano "peu", so it may be a variant form of the same code. | probable |
| 48, 4θ, 4r | – | – | Do not occur in this letter. | – |
| so../lo../do.. final pair | – | "loro" is spelled here `me n u k n` (l.11, l.12), i.e. L-O-[u]-[k]-O with two R-type signs | The Landriano pair `y Y'` is not in this letter. "loro" has an extra sign here too, which supports reading Landriano lo[y Y'] as "loro" with a doubled/variant sign. | guess |
| overbar on "∂ d p n f a ψ θ ψ 2" | l.3 | "de Montigia(?)" | Plain letters under a bar, probably a proper name spelled out. Not identified. | open |

## Glyph count (by eye, first pass)

About 1000 cipher glyphs in 20 lines (roughly 50 per line). About 80% have a value that makes a word in the reading above; about
10% have values but the sense is open (lines 2, 5, 8, 9, 13, 17, 18); about 10% (about 100) are unread, including the codes 42, rr ×3,
qq ×2 and ep. These are estimates, not counts: the tokens above are not complete enough to count exactly.

fr3045_f28_v2_reading.md

<!-- Reading of BnF fr. 3045 f. 28 (R4230), Claude Opus 5.5, 23 Oct 2026 (verify pass v8): the "Italian text" of fr3045_f28_final.md rewritten with that file's push-pass readings (l.5 'in fora, come batale perduta'; l.20 'che, obstante ogni mio mal tratamento'; l.2 'dico che lui proveda'; l.11/12 'loro'), as the 22 Oct review asked. Tokens fr3045_f28_tokens.txt; key KEY.md (Tomokiyo's Visconti cipher, extended). Code groups with no key kept as written: [42], [rr], [qq], [ep] (holes); [?word] = a reading the final file marks (?) or 'sense open'; [?] = unread; letters supplied inside a word in [ ]. Spelling as the signs give it (Lombard, 1528); punctuation added. fr3045_f28_reading.md is the first full pass, line by line. -->

# Galeazzo Visconti to Angelo Bolano, Alessandria, 15 Oct 1528 (BnF fr. 3045 f. 28): reading

## Reading
Bolano. [42] Solo ve dà questa [rr]: Savona che è stata socorsa, per la tarditate dil mondo; dico che lui proveda a tuto [qq], altramente tuto irà sempre male, de l'andata de Montigia per intrare in Savona; sep[a]r[at]ione sono causa si questo capo è stato de la reputacione in fora, come batale perduta, ogniuno voleva andare ale stancie, ch'era a disperare li coligati, quali poco mancano eserlo, et pur [?al] tuto tanto agio, dito, trovando espedienti che la cosa, che disperata, lo [?a] nocamento se harà cum Urbino; et se vederà reconciare almanco male cosa durisima. Tuti li aventureri partiti per loro poltronia et ingani de loro capitanei, [?quanto] [?ci] [?è] uno soldo, il che causa che Italiani che le hano [rr] li denari de nansi sono in Puglia. Designati [?qua] [rr] se mandava in as[?]sedio [?] [?havese] rimostrato la ruina [qq]; li lance che ne [?]convengano sono pochi, et serano asai tardi, e [?a] penso chi andrano paga gionti a [?]; Dio scia come sono malo contenti per servicio di [ep], che, obstante ogni mio mal tratamento, sarà la fine de la mia, cum ricomandarme ala sua bona gra[tia]. Da Alexandria ali XV de ottobre 1528.

## Score

`research/reading_audit/v2_score.py fr3045_f28_v2_reading.md it`: explicit holes 0.778 (14/18), plain 0.278 (the word list does not know 1528 Lombard spelling: dil, tarditate, ogniuno, stancie), hole rate 0.111, 198 words (23 Oct 2026). Sentences that do not build: 'et pur [?al] tuto tanto agio ... cum Urbino' (ll. 8-9), the 'quanto ci è' clause (ll. 12-13) and 'Designati ... as[?]sedio ... la ruina [qq]' (ll. 14-15).

fr3045_f28_en.md

# Galeazzo Visconti to Angelo Bolano, Alessandria, 15 Oct 1528 (BnF fr. 3045 f. 28): English translation

Translated from `fr3045_f28_v2_reading.md` (Claude Opus 5.5, 23 Oct 2026): the cipher, read with Tomokiyo's Visconti
key as extended in `KEY.md`. Holes as in the reading: `[42]`, `[rr]`, `[qq]`, `[ep]` code groups with no key value;
`[?word]` a doubtful reading; `[?]` unread; letters supplied inside a word in [ ]. The 1528 Lombard is often elliptic;
where the Italian does not build a sentence (the reading marks ll. 8-9, the "quanto ci è" clause and ll. 14-15), the
English follows it word for word and stays broken.

## Translation

Bolano. [42] This [rr] alone gives you: Savona, which has been relieved, for the slowness of the world; I say that he
should provide for everything [qq], otherwise everything will always go badly, of the going of Montigia to enter
Savona; divisions are the cause that this command has been, apart from its reputation, like a lost battle: everyone
wanted to go into winter quarters, so that it was enough to make the allies despair, who are not far from being so,
and yet [?at] all so much ease, I say, finding expedients so that the matter, which [was] desperate, the [?] harm
will be had with Urbino; and one will see a very hard thing mended, at least badly. All the adventurers [have]
left, through their cowardice and the deceits of their captains, [?as] [?there] [?is] one soldo, which is the cause
that the Italians who have [rr] them the advance money are in Puglia. Designated [?here] [rr] was being sent to the
si[?]ege [?] [?had] shown the ruin [qq]; the lances that [?]suit us are few, and will be very late, and [?I] think
who will go paid [when] arrived at [?]; God knows how ill content they are with the service of [ep], which,
notwithstanding all my ill treatment, will be the end of mine, commending myself to your good gra[ce]. From
Alessandria, 15 October 1528.

fr3096_f91_tokens.txt

# BnF fr. 3096 f.91r, Visconti to Francis I, pieve de Locate, June 1529. Glyph tokens, second pass (image b2 crops).
# One line per manuscript cipher line (L01-L19). Word groups separated by " | ".
# token=VALUE   read, part of a word that reads as sense
# token=VALUE?  value read, sense open (word not established)
# token=?       no value (unread)
# token=[code]  nomenclator code group, counted as one glyph
# Λ alone = doubling mark for the next letter (value written "+X"), except where marked E.
# The struck-through group in L19 (y y n) is cancelled in the manuscript and is not counted.
L01: ψ=I mp=L | a=T ψ=I p=M n=O k=R cΛ=E | b=H 7=A fn=NO | s=V cΛ=E f=N cΛ=E a=T ψ=I 7=A y=N ψ=I | ∂=D ψ=I | mp=L z÷=A | o=P 7=A l=C cΛ=E | o=P k=R ψ=I
L02: f=N l=C ψ=I o=P 7=A mp=L p=M cΛ=E f=N a=T Λ̇=E | o=P cΛ=E k=R | f=N n=O y=N | cΛ=E ne=S cΛ=E k=R cΛ=E | 7=A ne=S a=T k=R cΛ=E a=T ψ=I | l=C n=O f=N a=T k=R 7=A | ψ=I mp=L | a=T s=U k=R l=C n=O
L03: ψ=I mp=L | l=C b=H cΛ=E | ρ=M 7=A ψ=I | q=F 7=A k=R z÷=A y=N n=O | r̄r̄=[rr] | cΛ=E k=R ψ=I | l=C s=U p=M | ψ=I mp=L | ∂=D s=U l=C ⁊=A | ∂=D s=U k=R y=B ψ=I f=N n=O | cΛ=E a=T | o=P k=R n=O s=V ψ=I ∂=D ψ=I
L04: a=T n=O k=R cΛ=E | q=F cΛ=E l=C cΛ=E | mp=L n=O | q=F ψ=I l=C ψ=I n=O | ∂=D cΛ=E | s=V n=O ne=S a=T k=R n=O | h=B n=O f=N n=O | ne=S cΛ=E k=R s=V ψ=I a=T n=O k=R cΛ=E | ψ=I f=N | z÷=A ne=S cΛ=E
L05: l=C s=U k=R 7=A k=R | mp=L cΛ=E ψ=I | far=CHE | p=M 7=A ψ=I | s=V n=O ne=S a=T k=R 7=A | p=M 7=A ψ=I cΛ=E ne=S a=T z÷=A | 7=A h=B 7=A f=N ∂=D n=O f=N cΛ=E k=R 7=A | mp=L z÷=A | ne=S
L06: s=U 7=A | ne=S ψ=I θ=G y=N n=O k=R ψ=I 7=A | cΛ=E a=T | z=A mp=L a=T k=R n=O a=T 7=A f=N a=T n=O | ∂=D ψ=I ne=S cΛ=E | ∂=D ψ=I mp=L | ∂=D s=U l=C 7=A | ∂=D ψ=I | p=M ψ=I mp=L
L07: 7=A fn=NO | ψ=I mp=L | τ=Q s=U 7=A mp=L z÷=A | θ=G k=R 7=A f=N | o=P cΛ=E y=N z÷=A | p=M 7=A ψ=I | ne=S ψ=I | 7=A ne=S cΛ=E l=C s=U k=R cΛ=E k=R z÷=A | cΛ=E a=T ψ=I z÷=A p=M
L08: l=C s=U p=M | mp=L 7=A f=N ∂=D 7=A a=T z÷=A | ∂=D ψ=I | 48=[48] | p=M 7=A f=N l=C n=O | mp=L ψ=I | ∂=D n=O ψ=I | ne=S ψ=I θ=G y=N n=O k=R ψ=I | o=P k=R n=O p=M cΛ=E a=T n=O
L09: τ=Q s=U cΛ=E Λ=+L mp=L z÷=A | k=R ψ=I p=M 7=A ne=S cΛ=E Λ=+R k=R n=O | 7=A ne=S z÷=A ψ=I | ne=S z÷=A ψ=T? ne=S q=F 7=A a=T ψ=I | r̄r̄=[rr] | ∂=D ψ=I k=R n=O | h=B cΛ=E f=N cΛ=E
L10: l=C b=H ψ=I n=O | q=F 7=A l=C ψ=I n=O | ∂=D cΛ=E | mp=L cΛ=E | o=P k=R 7=A a=T ψ=I l=C b=H cΛ=E | ∂=D 7=A | p=M cΛ=E | 7=A | y=B n=O y=N +=O | o=P k=R n=O o=P +=O ne=S ψ=I a=T n=O
L11: ∂=D cΛ=E | mp=L cΛ=E | τ=Q s=U 7=A mp=L cΛ=E | p=M 7=A ψ=I | l=C cΛ=E Λ=+S ne=S z÷=A k=R n=O | ne=S cΛ=E? ψ=I f=N | far=CHE | s=V ψ=I s=V z÷=A | f=N cΛ=E Λ=? | s=V n=O θ=G mp=L ψ=I | y=N n=O
L12: a=T 7=A l=C cΛ=E k=R cΛ=E | far=CHE | fr=FU n=O | y=N? ne=S? 7=A? f=N? a=T? cΛ=E? | l=C? b=H? cΛ=E? k=R? ψ=I? | ψ=I f=N | l=C n=O f=N ne=S ψ=I θ=G mp=L ψ=I n=O | o=P 7=A k=R cΛ=E Λ=+S ne=S cΛ=E
L13: l=C b=H cΛ=E | ne=S cΛ=E | q=F s=U n=O ne=S cΛ=E | l=C n=O p=M L=C mp=L s=U ne=S n=O | ∂=D cΛ=E | o=P k=R n=O s=V 7=A k=R cΛ=E | ∂=D cΛ=E | ne=S ρ=? q=F n=O k=R l=C ψ=I k=R cΛ=E | p=M ψ=I mp=L
L14: 7=A fn=NO | p=M ψ=I 7=A | h=O o=P ψ=I f=N ψ=I n=O f=N cΛ=E | cΛ=E | far=CHE | fr=FU? | mp=L ψ=I | q=F 7=A k=R z÷=A | o=P cΛ=E k=R | mp=L n=O k=R n=O | p=M 7=A f=N l=C 7=A p=M cΛ=E
L15: f=N a=T n=O | o=P n=O f=N cΛ=E y=N ∂=D n=O | 7=A | l=C ⊥=A p=M o=P n=O | ∇=A a=T s=U Λ=+T a=T cΛ=E | b=H n=O k=R cΛ=E | ∂=D ψ=I q=F ψ=I l=C s=U mp=L a=T 7=A a=T cΛ=E | o=P cΛ=E k=R | mp=L cΛ=E | ∂Λ̄=?
L16: τ=Q s=U 7=A mp=L cΛ=E | ∂=D s=U k=R 7=A fn=NO | far=CHE | y=N n=O ψ=I | ne=S cΛ=E ψ=I 7=A p=M n=O | τ=Q s=U cΛ=E Λ=+L mp=L ψ=I | ∂=D ψ=I l=C 7=A fn=NO | ∂=D cΛ=E | o=P k=R ψ=I p=M z÷=A
L17: ∂=D cΛ=E | f=N n=O y=N | q=F 7=A k=R ne=S ψ=I | ψ=I mp=L | l=C b=H cΛ=E | f=N n=O f=N | mp=L ψ=I | 7=A f=N ∂=D z÷=A k=R 7=A | q=F 7=A a=T n=O | ne=S cΛ=E Λ=? | q8=[48] | ṗ=N? cΛ=E?
L18: l=C k=R cΛ=E ∂=D cΛ=E | l=C n=O p=M x=E? | ne=S? cΛ=E? a=T? cΛ=E? Λ=+R? k=R? n=O? | q=F 7=A k=R 7=A | cΛ=E a=T | ψ=I me=L | a=T s=U a=T n=O | ne=S ψ=I k=R cΛ=E | Λ=E | 7=A | s=U y=N n=O | ρ=F ψ=I f=N cΛ=E
L19: f=N n=O f=N | h=B n=O f=N n=O

fr3096_f91_final.md

# BnF fr. 3096 f.91r: Galeazzo Visconti to Francis I, camp at the pieve de Locate, June 1529 (second pass)

Image: img/fr3096_f91_v94.jpg (Gallica btv1b9060015d view 94, right half). Deskewed by 1.6°. Measured line centres
(y px, deskewed right half): 736, 881, 1021, 1167, 1303, 1447, 1593, 1734, 1879, 2017, 2162, 2315, 2443, 2603, 2740,
2880, 3020, 3157, 3303. Crops at 1.5x, with 2-2.5x zooms of the open stretches, are in scratchpad b2/. Every line was
redone glyph by glyph. The token file is fr3096_f91_tokens.txt. The counts below come from a script over that file.

Date: the clear close reads "ali XV de Giugno 1529" on the first pass. Tomokiyo gives 12 June. This pass did not
re-examine the date, so it is still to settle. The DECODE record number (R4230 or R4256) was not checked on this pass.

## Findings of this pass

- **The bare Λ is a doubling mark.** It doubles the letter that follows: que[Λ]la = quella (L9), rimase[Λ]ro (L9),
  ce[Λ]sarò = cessarò (L11), pare[Λ]se = paresse (L12), tu[Λ]te = tutte (L15), que[Λ]li = quelli (L16). The dotted
  Λ̇ in "principalmente" (L2) and the free-standing Λ in L18 ("è") are E.
- **"far" is not always CHE.** In L6 and at the start of L15 the same-looking group is three letters, f a n = NTO:
  "altrotanto" (L6) and "mancame|nto" (L14-15). This removes the first pass's "altro(?)ta(?) che ho" and "che ponendo".
- **y = B in "bono proposito"** (L10), as in "d'Urbino". Elsewhere y = N.
- **ρ = F in "fine"** (L18). In L3 the ρ of "mai" reads M. The two uses stay separate.
- **h = O** in "opinione" (L14). This is the only place where it has that value. Elsewhere h = B.

## Italian text (clear parts in italics)

*Syre: Sapendo* il timore hano Venetiani di la pace, principalmente per non esere astreti contra il Turco, il che mai
farano, [rr] heri cum il duca d'Urbino et providitore fece l'oficio de vostro bono servitore, in asecurar lei che mai
Vostra Maiestà abandonerà la sua Signoria, et altrotanto dise dil duca di Milano, il quala [cum] gran pena mai si
asecurerà, etiam cum l'andata di [48]. Manco li doi signori, prometo quella, rimasero asai satisfati(?). [rr] dirò
bene ch'io facio de le pratiche da me a bono proposito, de le quale mai cessarò s[e]in che viva, né[Λ?] vogli no tacere
che fuo [......](?) [cherì](?) in consiglio paresse, che se fuose concluso de provare de s[ρ]forcire Milano. Mia
opinione è che fu(?) li farà per loro mancamento, ponendo a campo a tutte hore dificultate per le [..] quale durano,
che noi seiamo quelli dicano de prima. De non farsi il che, non li andarà: fato se[Λ?] [48] ne(?) crede com[e](?)
[se terrò](?) farà, et il tuto, Sire, è a uno fine non bono.
*Syre, io suplico V[ost]ra M[aes]tà ... Datum in campo in la pieve de Loccate ali XV de Giugno 1529.
Le vostre humile et obeyent servitor et subject Galeazo Vesconte.*

(The clear close is kept as the first pass read it. It was not re-read on this pass.)

## English summary

Visconti writes to the King from the League camp at the pieve de Locate in mid-June 1529. The Venetians fear the peace
(the Cambrai talks, with the Treaty of Barcelona just concluded between the Emperor and the Pope). Above all they fear
being bound to act against the Turk, "which they will never do". Yesterday [rr] did the office of the King's good
servant with the Duke of Urbino and the Venetian proveditor. He assured them that His Majesty will never abandon the
Signoria, and he said as much of the Duke of Milan. The Duke of Milan, Visconti adds, will hardly ever feel secure,
even with the arrival or departure ("l'andata") of [48]. Visconti assures the King that the two lords were well
satisfied all the same. [rr] will report favourably on the negotiations Visconti conducts, to good purpose. Visconti
will never stop them while he lives, and he will not keep silent. He had urged in council that, once terms were
concluded, the army should try to force Milan. In his opinion the others will fail through their own shortcomings,
putting difficulties in the way at every hour. Those difficulties last, and Visconti's party are the ones who spoke of
them first. If Milan is not attempted it will not happen. If [48] does not believe it, the outcome is foreseeable,
"and the whole thing, Sire, is heading to no good end." The letter was written days before Saint-Pol's defeat at
Landriano (21 June 1529).

## Code groups

| code | where | context | identification |
|---|---|---|---|
| [48] | L8, L17 (the L17 "q8" is the same group: a crossed 4 with a looped head, then 8) | "etiam cum l'andata di [48]"; "[48] ne crede" | Probably **Saint-Pol**, the King's lieutenant, who led the French army and was about to march toward Genoa. This also fits "a casa de [48]" in the 1528 letter. Not proved. |
| [rr] (overbarred) | L3, L9 | "[rr] heri ... fece l'oficio de vostro bono servitore"; "[rr] dirò bene ch'io facio de le pratiche" | A person working for the King with the Venetians: the French ambassador at Venice or with the Venetian army. It could also be a first-person form. "fece" can be a Lombard first person, but "dirò bene ch'io facio" argues against Visconti being rr. The identity is open. |

## Gaps (from the token file)

| line | glyphs | state | cause |
|---|---|---|---|
| L9 | 1 (ψ in "sa[ψ]isfati") | sense open | "satisfati" needs T, but the sign reads I. It is either a slip or a misread shape. |
| L11 | 1 (cΛ in "s[e]in") | sense open | "sin che viva" with an extra E. |
| L11 | 1 (Λ after "ne") | unread | This Λ doubles nothing before "vogli". It may be a false start. |
| L12 | 6 (y ne 7 f a cΛ, "[.]sante") | sense open | No word fits "che fuo [N]SANTE". "instante" and "presente" would need other signs. |
| L12 | 5 (l b cΛ k ψ, "cherì") | sense open | This could be archaic "cherì" (asked, from cherere), but that is not established. |
| L13 | 1 (ρ in "s[ρ]forcire") | unread | Probably an extra letter. |
| L14 | 1 ("fu" group) | sense open | "che fu li farà": the value is FU, but the syntax is unclear. |
| L15 | 1 (∂Λ̄ at the line end) | unread | An overbarred sign before "quale". It may be a code or a flourish. |
| L17 | 1 (Λ in "fato se[Λ]") | unread | A doubling mark before the code has no effect. |
| L17 | 2 (dotted p + E, "ne") | sense open | The dotted p is read N ("[48] ne crede"). That value is not confirmed elsewhere. |
| L18 | 1 (x in "com[x]") | sense open | "come" is likely. The sign is unique. |
| L18 | 7 ("se terrò") | sense open | The values S E T E +R R O are clear. The phrase does not resolve. |
| — | 4 code groups | code | [48] ×2, [rr] ×2: numbered nomenclator entries with no surviving key list (outside blocker). |

## Counts (script over fr3096_f91_tokens.txt; each code group = 1 glyph; the struck y y n in L19 excluded)

| total glyphs | read as sense | sense open | unread | codes |
|---|---|---|---|---|
| 699 | **667 (95.4%)** | 24 | 4 | 4 |

Excluding the code groups from the base, 667 / 695 = 96.0%. The 95% bar is reached, narrowly: 1 of every 21 glyphs is
still open, most of them in L12 and L18.

fr3096_f91_reading.md

# BnF fr. 3096 f.91r: Galeazzo Visconti to Francis I, camp at the pieve de Locate, 15 June 1529 (first pass)

Source: siblings/fr3096_f91_v94.jpg, right half (Gallica btv1b9060015d view 94). The letter is complete on f.91r:
19 cipher lines, then the clear close and the signature ("Le vostre humile et obeyent servitor et subject Galeazo
Vesconte"). View 95 (f.91v) was not needed. Crops: scratchpad a6/Z<line>_<quarter>.png. Measured line centres
(quarter 0, y px): 720, 865, 1004, 1141, 1276, 1409, 1546, 1674, 1813, 1943, 2101, 2230, 2379, 2491, 2636, 2799,
2947, 3098, 3257. The lines slope down to the right by about 30-45 px across the text block.

Status: **first pass, read in part.** About three quarters of the lines give running Italian. The rest has the values
read but the sense is open, or has uncertain shapes. Tokens were transcribed by eye from the native-resolution crops.

## Sign values used and added on this letter

Same system as fr. 3034 f.154 (KEY.md). New or confirmed on this letter:

| sign | value | evidence (line) |
|---|---|---|
| f (long s-like, no tail) | **N** (a second N) | Venetiani L1, hano L1, contra L2, principalmente L2, gran pena L7 |
| fn | NO (f + n) | hano L1, non L2, durano/dicano L16 |
| b (b-shape) | H | il che L3/L17, hano L1 |
| h (h-shape) | B | bono servitore L4, abandonerà L5, bono L19 |
| q (loop on a stem, "9") | **F** | facio L10, fece l'oficio L4, farà L3, fatose L17 |
| ∂ (θ/9-like round d) | D | di L1, duca L3, d'Urbino L3, durano L16 |
| cΛ, Λ, Λ̇ (d with rising tail, with or without bowl or dot) | E | timore L1, principalmente L2, esere L2 |
| z with two dots (z÷) | A | di la L1, Maiestà L5, etiam L7, andata L8 |
| τ (hooked tau) | Q | quala L7, quale L11 and L16, queli L16 |
| y bent (in d'Urbino) | B | d'Urbino L3 |
| θ | G | Signoria L6, gran L7, signori L8, soglio L11-12, consiglio L12 |
| 4 (open figure 4) + 8 | code [48] | L8 |
| overbarred rr | code [rr] | L3, L9 |

Thin diagonal strokes and long horizontal bars through descenders are later dividers (as on 3034). They were ignored.

## Lines (tokens; Italian)

`[..]` = no value; `(?)` = values read, sense open.

| L | tokens (abridged where a stretch was unreadable) | Italian |
|---|---|---|
| clear | Syre: Sapendo | Syre: Sapendo |
| 1 | ψ mp · a ψ p n k cΛ · b 7 fn · s cΛ f cΛ a ψ 7 y ψ · ∂ ψ mp z÷ · o 7 l cΛ · o k ψ | il timore hano Venetiani di la pace, pri- |
| 2 | f l ψ o 7 mp p cΛ f a Λ̇ · o cΛ k · f n y · cΛ ne cΛ k cΛ · 7 ne a k cΛ a ψ · l n f a k 7 · ψ mp · a s k l n | -ncipalmente per non esere astreti contra il Turco. |
| 3 | ψ mp l b cΛ · ρ 7 ψ · q 7 k ⁊ · y n · **[r̄r̄]** · cΛ k ψ · l s p · ψ mp · ∂ s l ⁊ · ∂ s k y ψ f n · cΛ a · o k n s ψ ∂ ψ | Il che mai farà no(?). [rr] heri cum il duca d'Urbino et providi- |
| 4 | a n k cΛ · q cΛ · l cΛ · mp n · q ψ l ψ n · ∂ cΛ · s n ne a k n · h n f n · ne cΛ k s ψ a n k cΛ · ψ f z÷ ne cΛ | -tore fece l'oficio de vostro bono servitore, in a se(?) |
| 5 | l s k 7 k mp cΛ ψ · far · p 7 ψ · s n ne a k 7 · p 7 ψ cΛ ne a z÷ · 7 h 7 f ∂ n f cΛ k 7 · mp z÷ ne | curarli(?) che mai Vostra Maiestà abandonerà la s- |
| 6 | s 7 · ne ψ θ y n k ψ 7 · cΛ a · z mp a k n a 7 · far · n · ∂ ψ ne cΛ · ∂ ψ mp · ∂ s l 7 · ∂ ψ · p ψ mp | -ua Signoria, et altro(?)ta(?) che ho dise dil duca di Mil- |
| 7 | 7 fn · ψ mp · τ s 7 mp z÷ · θ k 7 f · o cΛ y z÷ · p 7 ψ · ne ψ · 7 ne cΛ l s k cΛ k z· · cΛ a ψ z÷ p | -ano, il quala [cum] gran pena mai si asecurerà, etiam |
| 8 | l s p · mp 7 f ∂ 7 a z÷ · ∂ ψ · **[4 8]** · p 7 f l n · mp ψ · ∂ n ψ · ne ψ θ y n k ψ · o k n p cΛ a n | cum l'andata di [48], manco li doi signori prometon(?)- |
| 9 | τ s cΛ Λ mp z÷ · k ψ p 7 ne cΛ Λ k n · 7 ne z÷ ψ · ne ·a ψ ne q 7 a ψ(?) · **[r̄r̄]** · ∂ ψ k n · h cΛ f cΛ | quela rimasero asai [..](?); [rr] dirò bene |
| 10 | l b ψ n · q 7 l ψ n · ∂ cΛ · mp cΛ · o k 7 a ψ l b cΛ · ∂ 7 p cΛ 7 · y n y + · o k n o + ne ψ a n | ch'io facio de le pratiche da me a non a proposito(?) |
| 11 | ∂ cΛ mp cΛ · τ s 7 mp cΛ · p 7 ψ · l cΛ Λ ne z÷ k n · ne cΛ ψ f · far · s ψ s z÷ · f cΛ Λ s n · ne n θ mp ψ | de le quale mai cesarò(?) se in(?) che viva(?) ne[..]o(?) soglio |
| 12 | y n · a 7 l cΛ k cΛ · far · fr n · y ne 7 q a cΛ(?) · l b cΛ k ψ · ψ f(?) · l n f ne ψ θ mp ψ n · o 7 k cΛ Λ ne cΛ | no tacere, che fuo [..](?) che ri(?) in consiglio parese |
| 13 | l b cΛ · ne cΛ · q s n ne cΛ · l n p L mp ne n ne n(?) · ∂ cΛ · o k n s 7 k cΛ · ∂ cΛ · ne ρ q n k l ψ k cΛ · p ψ mp | che se fuose com[..](?), de provare de sforcire(?) Mil- |
| 14 | 7 fn · p ψ 7 · h o ψ f ψ n f cΛ · cΛ · far · fr n · mp ψ · ψ q 7 k z÷ · o cΛ k · mp n k n · p 7 f l 7 p cΛ | -ano: mia opinione è che fuo li farà(?) per loro mancame- |
| 15 | far · o n f cΛ y ∂ n · 7 · l ⊥ p n · 7 a s cΛ cΛ h n k cΛ(?) · ∂ ψ q ψ l s mp a 7 a cΛ · o cΛ k mp o cΛ | che ponendo a campo(?) [..](?) dificultate per l'o(?)pe- |
| 16 | τ s 7 mp cΛ · ∂ s k 7 fn · far · y n ψ · ne cΛ ψ 7 p n · τ s cΛ mp ψ · ∂ ψ l 7 fn · ∂ cΛ · o k ψ p z÷ | quale durano, che noi seiamo queli dicano de prima. |
| 17 | ∂ cΛ f n y q 7 k ne ψ · ψ mp l b cΛ · f n f mp ψ · ψ ∂ 7 f ∂ z÷ k 7 · q 7 a n ne cΛ Λ · **[q 8]**(?) · p ψ cΛ | De non farsi il che, non li [..]andarà(?) fatose [q8](?) mie(?) |
| 18 | l k cΛ ∂ cΛ · l n p x · ne ρ a cΛ Λ k n · q 7 k 7 · cΛ a · ψ mp cΛ a s 7 f ne cΛ ψ k cΛ · cΛ Λ τ 7 · s y n ρ ψ f cΛ | crede com[a](?) st[e]ro(?) farà, et il [..](?) [..] fine(?) |
| 19 | f n f · (y y n struck through) · h n f n | non bono. |
| clear | Syre, io suplico V[ost]ra M[aes]tà ... Datum in campo in la pieve de Loccate ali XV de Giugno 1529 | |

### Running text (what is secure)

Syre: sapendo il timore hano Venetiani di la pace, principalmente per non esere astreti contra il Turco, il che mai
farà no(?). [rr] heri cum il duca d'Urbino et providitore fece l'oficio de vostro bono servitore ... che mai Vostra
Maiestà abandonerà la sua Signoria ... dil duca di Milano, il quala [cum] gran pena mai si asecurerà, etiam cum
l'andata di [48], manco li doi signori ... [rr] dirò bene ch'io facio de le pratiche ... soglio no tacere ... in
consiglio parese ... de provare de sforcire(?) Milano: mia opinione è ... per loro mancamento(?) ... quale durano, che
noi seiamo queli dicano de prima. De non farsi il che ... non bono.

English: knowing the Venetians' fear of the peace (the talks leading to Cambrai), above all of being bound to act
against the Turk, [rr] did the office of the King's good servant yesterday with the Duke of Urbino and the
proveditor, [assuring them] that His Majesty will never abandon the Signoria. The Duke of Milan will hardly ever be
secured, "even with the departure of [48]". Visconti says he does not keep silent in council and gives his opinion on
attempting Milan.

## Code groups in this letter

| code | line | context | identification | strength |
|---|---|---|---|---|
| **48** (open 4 + 8) | L8 | "etiam cum l'andata di [48]": the Duke of Milan will hardly be secured, even with 48's departure | **Saint-Pol** (François de Bourbon, comte de Saint-Pol, the King's lieutenant in Italy). In mid-June 1529 he was leaving the League camp for Genoa, which led to Landriano on 21 June. It also fits the 1528 letter, where the council met "a casa de [48]" in the French camp at Landriano and the Duke of Urbino came to it. | probable (fits both contexts; no clear-text equation) |
| q8 (?) | L17 end | "fatose [q8] mie(?)"; the first sign is a looped q crossed by a divider, not the open 4 of L8 | possibly a second 48 written with a q-like 4, or the letters F + 8 | guess |
| **rr** (overbarred) | L3, L9 | L3: "[rr] heri cum il duca d'Urbino et providitore fece l'oficio de vostro bono servitore"; L9: "[rr] dirò bene ch'io facio de le pratiche" | A person code for someone acting for the King with Urbino and the Venetian proveditor. In L3 the verb is third person (fece), so rr is not the King and probably not Visconti. If 48 is Saint-Pol, rr is another French agent: the ambassador to Venice, or a French captain in the camp. If 48 is not Saint-Pol, rr is Saint-Pol. The overbar marks it as a nomenclator entry, not two letters. | guess (a person working for the King: probable) |
| 4θ, 4r, qq, peu | — | not found in this letter | — | — |
| the so../lo../do.. final pair (y Y') | — | not found | — | — |

The King as "peu": the letter is addressed to Francis I, and he appears only as "Vostra Maiestà", spelled out in
cipher letters (L5 `s n ne a k 7 · p 7 ψ cΛ ne a z÷`), and "Syre" in clear. That fits "peu" (twice in the 1528 letter,
"scriva al peu", "servicio dil peu") being the King, but this letter cannot prove it. **peu = King: probable from
the 1528 contexts, not tested here.**

Consequence for the 1528 letter: "a casa de [48]" = at Saint-Pol's lodging (probable). The r21 [r r ᵽ] may be the
code rr followed by one sign. It is not the King, because rr stands for a third person working for the King.

## Glyph counts (by eye, this pass)

| lines | cipher glyphs (approx.) | values assigned | sense established |
|---|---|---|---|
| L1-L19 | about 590 | about 560 (95%) | about 430 (73%) |

These counts are estimates from the token rows above, not machine counts. Code groups (48, rr ×2, q8) count as one
glyph each.

## Open

- L9 middle, L11 middle, L12 "fuo [..]", L13 "com[..]", L15 "a [..]", L17-18: the values are read but the words do
  not resolve. A second pass with 3x crops of these stretches is needed.
- It is still to settle whether the bare Λ is always E or sometimes a doubling or other sign. Compare "quela" L9,
  "cesarò" L11, "parese" L12.
- h (B) and b (H) are separate signs. The O-like h in L14 "opinione" needs a check.

fr3096_f91_en.md

<!-- English translation of the Italian text of fr3096_f91_final.md (Claude Opus 5.5, 23 Oct 2026): BnF fr. 3096 f. 91r (R4256), the second-pass reading; the clear parts in *italics* as there. Same marks: [rr] (the writer's code name), [48] (a person), [cum] a code read as "with"; [..] and [......] unread; (?) doubtful; [Λ?], [ρ] doubtful signs. -->

# Galeazzo Visconti to Francis I, camp at the pieve de Locate, 15 June 1529 (BnF fr. 3096 f. 91r): English translation

*Sire: Knowing* the fear the Venetians have of the peace, chiefly so as not to be bound against the Turk, which they will
never do, [rr] yesterday, with the Duke of Urbino and the provveditore, did the office of your good servant in assuring
them that Your Majesty will never abandon their Signoria; and as much he said of the Duke of Milan, who will never, [with]
great pain, be reassured, even with the going of [48]. At least the two lords, I promise you, remained well satisfied(?).
[rr] I will say indeed that I carry on dealings on my own to good purpose, from which I shall never cease as long as I
live; nor do I wish to keep silent that it was [......](?) [cherì](?) seemed good in council, that if it had been
concluded to try to s[ρ]force Milan. My opinion is that it will(?) be done for them through their own failing, raising
difficulties in the field at all hours, for the [..] which last, that we are those who speak first. If it is not done,
it will not go there: done se[Λ?] [48] ne(?) believes how(?) [I shall hold](?) it will do, and the whole thing, Sire, is
to no good end.

*Sire, I beg Your Majesty ... Given in camp at the pieve de Locate, the 15th of June 1529. Your humble and obedient
servant and subject Galeazo Vesconte.*