Let's repeat the investigation of [czHP] strings, but including the `8' letter:
cat bio-j-jsa-gut.wds \
| sed \
-e 's/^/_/g' \
-e 's/$/_/g' \
-e 's/[ql]j/H/' \
-e 's/[ql]g/P/' \
-e 's/cs/z/g' \
-e 's/ij/k/g' \
-e 's/ix/e/g' \
-e 's/is/r/g' \
-e 's/iiu/n/g' \
-e 's/y/i/g' \
-e 's/ci/a/g' \
-e 's/cg/8/g' \
| enum-contexts -vPAT='[czHP8][czHP8]*' -vCTX=0 \
| wfreq
793 0.19 H
382 0.09 8
374 0.09 Hc8
314 0.07 Hcc8
305 0.07 ccc8
277 0.06 zcc8
178 0.04 Hcc
163 0.04 ccc
152 0.04 z
140 0.03 zcc
102 0.02 Hc
74 0.02 cc
56 0.01 cccHc
49 0.01 zccc8
49 0.01 Pccc8
49 0.01 P
48 0.01 cc8
46 0.01 zc
41 0.01 ccccHc
40 0.01 cccc
39 0.01 zccHc
35 0.01 zcccHc
35 0.01 Hccc8
34 0.01 zccc
27 0.01 cccc8
25 0.01 Hccc
24 0.01 zccH
20 0.00 zc8
18 0.00 cccH
18 0.00 8zcc8
16 0.00 Hzcc
15 0.00 Hzcc8
14 0.00 cHc
14 0.00 Pccc
13 0.00 zcccH
12 0.00 ccccH
12 0.00 cHcc
11 0.00 cccz
11 0.00 ccH
11 0.00 8ccc8
9 0.00 zccHcc
9 0.00 cccHc8
8 0.00 cHcc8
7 0.00 cccHcc8
7 0.00 Pzcc8
7 0.00 Hzc8
6 0.00 zccHcc8
6 0.00 8cc
5 0.00 zcccHcc8
5 0.00 zcH
5 0.00 ccccz
5 0.00 cHc8
5 0.00 c
5 0.00 Pcc8
5 0.00 8cc8
4 0.00 zzcc8
4 0.00 ccccHcc
4 0.00 cccHcc
4 0.00 cH
4 0.00 Pcc
4 0.00 Hcccc
4 0.00 8ccc
3 0.00 zcccz
3 0.00 cccP
3 0.00 ccHc8
3 0.00 cPcc
3 0.00 cPc
3 0.00 P8
3 0.00 8zcc
3 0.00 8zc
2 0.00 zzcc
2 0.00 zcccPc
2 0.00 zcccHcc
2 0.00 zcccHc8
2 0.00 zccPcc
2 0.00 zccHc8
2 0.00 ccz
2 0.00 ccccHcc8
2 0.00 cccPcc8
2 0.00 cccPcc
2 0.00 cP
2 0.00 Pzc8
2 0.00 Pzc
2 0.00 Pcccc
2 0.00 Hzc
2 0.00 Hczcc
2 0.00 Hczc
2 0.00 Hccz
2 0.00 Hcccc8
2 0.00 H8
2 0.00 8zccc8
2 0.00 8zccc
2 0.00 8cccc
2 0.00 8c8
2 0.00 8Hc8
1 0.00 zzcccHc
1 0.00 zzccH
1 0.00 zzcHcc8
1 0.00 zcz8
1 0.00 zccz
1 0.00 zccccHcc
1 0.00 zcccc
1 0.00 zcccHcc8cc
1 0.00 zccPccc8
1 0.00 zccP
1 0.00 zccHccc
1 0.00 zcHcc
1 0.00 zcHc
1 0.00 zPcc
1 0.00 zHcc
1 0.00 zH
1 0.00 ccccc
1 0.00 ccccPcc8
1 0.00 cccPccc8
1 0.00 cccHccc8
1 0.00 ccc8cc
1 0.00 ccPzccc8
1 0.00 ccPzccc
1 0.00 ccPccc8
1 0.00 ccP
1 0.00 ccHcc8
1 0.00 ccHcc
1 0.00 cc8cc
1 0.00 cPccc8
1 0.00 cPccc
1 0.00 cPcc8
1 0.00 cPc8
1 0.00 cHccz
1 0.00 cHccc8
1 0.00 Pzcc
1 0.00 Hczc8
1 0.00 Hcz8
1 0.00 Hcz
1 0.00 Hccz8
1 0.00 Hc8zcc8
1 0.00 Hc8cc
1 0.00 Hc8c8
1 0.00 Hc8c
1 0.00 8zcccz
1 0.00 8zc8
1 0.00 8cccc8
1 0.00 8cccHcc8
1 0.00 8Hzcc
1 0.00 8Hcc
1 0.00 88
----- ---- ----
4282 1.00 TOT
Apparently the `8' (\cg/) does not tend to be surrounded by [czHP] strokes, it is either preceded
or followed by them. Thus `8' seems quite unlike `P'.
Let's look at some `P' strings and try to find similar words with the `P' replaced by something else:
cat bio-j-jsa-gut.wds \
| sed \
-e 's/^/_/g' \
-e 's/$/_/g' \
-e 's/[ql]j/H/' \
-e 's/[ql]g/P/' \
-e 's/cs/z/g' \
-e 's/ij/k/g' \
-e 's/ix/e/g' \
-e 's/is/r/g' \
-e 's/iiu/n/g' \
-e 's/y/i/g' \
-e 's/ci/a/g' \
-e 's/cg/8/g' \
| egrep '[^czHP]P[^czHP]' \
| wfreq
8 0.15 _Poe_
2 0.04 _oPoea_
2 0.04 _oPar_
2 0.04 _oPa_
2 0.04 _Poeccc8a_
1 0.02 _qoPor_
1 0.02 _qoPoe_
1 0.02 _qoPa_
1 0.02 _qoP_
1 0.02 _qoP8a_
1 0.02 _qoHoPa_
1 0.02 _oePocHca_
1 0.02 _oPor_
1 0.02 _oPoe_
1 0.02 _oPaezcc8a_
1 0.02 _oPaea_
1 0.02 _oPae_
1 0.02 _oPaeHain_
1 0.02 _ePoe_
1 0.02 _Poin_
1 0.02 _Poezcca_
1 0.02 _Poezca_
1 0.02 _Poezc8ae_
1 0.02 _Poezc8a_
1 0.02 _Poeccc8_
1 0.02 _Poecca_
1 0.02 _Poecc8arcn_
1 0.02 _Poecc8a_
1 0.02 _Poearar_
1 0.02 _Poeain_
1 0.02 _Poeaecc8a_
1 0.02 _PoeHczcoe_
1 0.02 _PoeHcca_
1 0.02 _Poe8zcc8a_
1 0.02 _Poe8aHa_
1 0.02 _PoHan_
1 0.02 _Par_
1 0.02 _PaHc8a_
1 0.02 _P8oe_
1 0.02 _P8aezcor_
1 0.02 _HoePa_
----- ---- ----
52 1.00 TOT
set noglob
foreach f ( \
'_'.'oe_' \
'_o'.'oea_' \
'_o'.'ar_' \
'_o'.'a_' \
'_'.'oeccc8a_' \
)
echo " "
echo "-----------------------------------------------------------------------"
echo " "
cat bio-j-jsa-gut.wds \
| sed \
-e 's/^/_/g' \
-e 's/$/_/g' \
-e 's/[ql]j/H/' \
-e 's/[ql]g/P/' \
-e 's/cs/z/g' \
-e 's/ij/k/g' \
-e 's/ix/e/g' \
-e 's/is/r/g' \
-e 's/iiu/n/g' \
-e 's/y/i/g' \
-e 's/ci/a/g' \
-e 's/cg/8/g' \
| compare-contexts -rctx 0 -lctx 0 -colw 24 \
"${f:r}P${f:e}" "${f:r}[^P]${f:e}" "${f:r}[^P][^P]${f:e}"
end
unset noglob
-----------------------------------------------------------------------
8 1.00 _Poe_ 81 0.52 _qoe_ 11 0.19 _zcoe_
----- ---- ---- 25 0.16 _zoe_ 11 0.19 _oHoe_
8 1.00 TOT 17 0.11 _eoe_ 11 0.19 _ccoe_
17 0.11 _8oe_ 8 0.14 _oeoe_
9 0.06 _roe_ 5 0.09 _oroe_
6 0.04 _Hoe_ 3 0.05 _Hcoe_
----- ---- ---- 2 0.04 _aroe_
155 1.00 TOT 1 0.02 _qooe_
1 0.02 _oqoe_
1 0.02 _eHoe_
1 0.02 _e8oe_
1 0.02 _aeoe_
1 0.02 _8roe_
----- ---- ----
57 1.00 TOT
-----------------------------------------------------------------------
2 1.00 _oPoea_ 2 1.00 _oroea_ ----- ---- ----
----- ---- ---- ----- ---- ---- 0 1.00 TOT
2 1.00 TOT 2 1.00 TOT
-----------------------------------------------------------------------
2 1.00 _oPar_ 35 0.92 _oHar_ 7 1.00 _oeHar_
----- ---- ---- 1 0.03 _orar_ ----- ---- ----
2 1.00 TOT 1 0.03 _oear_ 7 1.00 TOT
1 0.03 _o8ar_
----- ---- ----
38 1.00 TOT
-----------------------------------------------------------------------
2 1.00 _oPa_ 25 0.45 _oHa_ 21 0.53 _oHca_
----- ---- ---- 23 0.41 _oea_ 10 0.25 _oeHa_
2 1.00 TOT 6 0.11 _ora_ 9 0.23 _oe8a_
2 0.04 _o8a_ ----- ---- ----
----- ---- ---- 40 1.00 TOT
56 1.00 TOT
-----------------------------------------------------------------------
2 1.00 _Poeccc8a_ 6 0.67 _qoeccc8a_ ----- ---- ----
----- ---- ---- 2 0.22 _zoeccc8a_ 0 1.00 TOT
2 1.00 TOT 1 0.11 _8oeccc8a_
----- ---- ----
9 1.00 TOT
-----------------------------------------------------------------------
It ssems that isolated `P' = {\lg/,\qg/} is closely related to `r'=\is/,
`q' = \q/, `z' = \cs/, `8' = \cg/, `e' = \ix/, `H' = {\lj/,\qj/}.
cat bio-j-jsa-gut.wds \
| sed \
-e 's/^/_/g' \
-e 's/$/_/g' \
-e 's/[ql]j/H/' \
-e 's/[ql]g/P/' \
-e 's/cs/z/g' \
-e 's/ij/k/g' \
-e 's/ix/e/g' \
-e 's/is/r/g' \
-e 's/iiu/n/g' \
-e 's/y/i/g' \
-e 's/ci/a/g' \
-e 's/cg/8/g' \
| egrep '[^czHP]Pccc[^czHP]' \
| wfreq
14 0.22 _oPccc8a_
14 0.22 _Pccc8a_
8 0.13 _qoPccc8a_
4 0.06 _oePccc8a_
4 0.06 _oPccca_
4 0.06 _Pcccoe_
4 0.06 _Pccc8ar_
3 0.05 _Pccca_
2 0.03 _qoPccca_
1 0.02 _oqoPccc8a_
1 0.02 _ePccc8a_
1 0.02 _aPccca_
1 0.02 _aPccc8a_
1 0.02 _Pccc8ae_
1 0.02 _8oePccc8a_
----- ---- ----
63 1.00 TOT
set noglob
foreach f ( \
'_o'.'ccc8a_' \
'_'.'ccc8a_' \
'_qo'.'ccc8a_' \
'_oe'.'ccc8a_' \
'_o'.'ccca_' \
'_'.'cccoe_' \
'_'.'ccc8ar_' \
'_'.'ccca_' \
'_qo'.'ccca_' \
)
echo " "
echo "-----------------------------------------------------------------------"
echo " "
cat bio-j-jsa-gut.wds \
| sed \
-e 's/^/_/g' \
-e 's/$/_/g' \
-e 's/[ql]j/H/' \
-e 's/[ql]g/P/' \
-e 's/cs/z/g' \
-e 's/ij/k/g' \
-e 's/ix/e/g' \
-e 's/is/r/g' \
-e 's/iiu/n/g' \
-e 's/y/i/g' \
-e 's/ci/a/g' \
-e 's/cg/8/g' \
| compare-contexts -rctx 0 -lctx 0 -colw 24 \
"${f:r}P${f:e}" "${f:r}[^P]${f:e}" "${f:r}[^P][^P]${f:e}"
end
unset noglob
-----------------------------------------------------------------------
14 1.00 _oPccc8a_ 23 0.79 _oeccc8a_ 1 0.50 _orzccc8a_
----- ---- ---- 5 0.17 _oHccc8a_ 1 0.50 _oecccc8a_
14 1.00 TOT 1 0.03 _o8ccc8a_ ----- ---- ----
----- ---- ---- 2 1.00 TOT
29 1.00 TOT
-----------------------------------------------------------------------
14 1.00 _Pccc8a_ 52 0.37 _eccc8a_ 23 0.49 _oeccc8a_
----- ---- ---- 36 0.26 _zccc8a_ 5 0.11 _oHccc8a_
14 1.00 TOT 19 0.13 _cccc8a_ 4 0.09 _ezccc8a_
14 0.10 _Hccc8a_ 2 0.04 _rzccc8a_
9 0.06 _8ccc8a_ 2 0.04 _qoccc8a_
5 0.04 _rccc8a_ 2 0.04 _ecccc8a_
4 0.03 _accc8a_ 2 0.04 _eHccc8a_
1 0.01 _qccc8a_ 1 0.02 _o8ccc8a_
1 0.01 _occc8a_ 1 0.02 _eoccc8a_
----- ---- ---- 1 0.02 _azccc8a_
141 1.00 TOT 1 0.02 _aeccc8a_
1 0.02 _acccc8a_
1 0.02 _8zccc8a_
1 0.02 _8cccc8a_
----- ---- ----
47 1.00 TOT
-----------------------------------------------------------------------
8 1.00 _qoPccc8a_ 11 0.65 _qoHccc8a_ 2 0.40 _qoezccc8a_
----- ---- ---- 6 0.35 _qoeccc8a_ 2 0.40 _qoHcccc8a_
8 1.00 TOT ----- ---- ---- 1 0.20 _qoecccc8a_
17 1.00 TOT ----- ---- ----
5 1.00 TOT
-----------------------------------------------------------------------
4 1.00 _oePccc8a_ 1 1.00 _oecccc8a_ 1 1.00 _oeoHccc8a_
----- ---- ---- ----- ---- ---- ----- ---- ----
4 1.00 TOT 1 1.00 TOT 1 1.00 TOT
-----------------------------------------------------------------------
4 1.00 _oPccca_ 12 0.63 _oeccca_ 2 0.40 _oezccca_
----- ---- ---- 6 0.32 _oHccca_ 2 0.40 _oecccca_
4 1.00 TOT 1 0.05 _orccca_ 1 0.20 _oeHccca_
----- ---- ---- ----- ---- ----
19 1.00 TOT 5 1.00 TOT
-----------------------------------------------------------------------
4 1.00 _Pcccoe_ 3 0.27 _zcccoe_ 1 0.33 _oecccoe_
----- ---- ---- 3 0.27 _8cccoe_ 1 0.33 _cccccoe_
4 1.00 TOT 2 0.18 _ecccoe_ 1 0.33 _8ccccoe_
2 0.18 _Hcccoe_ ----- ---- ----
1 0.09 _acccoe_ 3 1.00 TOT
----- ---- ----
11 1.00 TOT
-----------------------------------------------------------------------
4 1.00 _Pccc8ar_ 1 0.25 _eccc8ar_ ----- ---- ----
----- ---- ---- 1 0.25 _cccc8ar_ 0 1.00 TOT
4 1.00 TOT 1 0.25 _accc8ar_
1 0.25 _Hccc8ar_
----- ---- ----
4 1.00 TOT
-----------------------------------------------------------------------
3 1.00 _Pccca_ 31 0.39 _cccca_ 12 0.35 _oeccca_
----- ---- ---- 23 0.29 _zccca_ 6 0.18 _oHccca_
3 1.00 TOT 15 0.19 _eccca_ 3 0.09 _ezccca_
5 0.06 _Hccca_ 3 0.09 _azccca_
4 0.05 _rccca_ 2 0.06 _ecccca_
1 0.01 _accca_ 2 0.06 _8zccca_
1 0.01 _8ccca_ 1 0.03 _rcccca_
----- ---- ---- 1 0.03 _qoccca_
80 1.00 TOT 1 0.03 _orccca_
1 0.03 _acccca_
1 0.03 _Hcccca_
1 0.03 _8cccca_
----- ---- ----
34 1.00 TOT
-----------------------------------------------------------------------
2 1.00 _qoPccca_ 8 0.50 _qoeccca_ 1 0.50 _qoeHccca_
----- ---- ---- 8 0.50 _qoHccca_ 1 0.50 _qoHcccca_
2 1.00 TOT ----- ---- ---- ----- ---- ----
16 1.00 TOT 2 1.00 TOT
It seems that `Pccc' is closely related to `Hccc' `eccc' `zccc' `8ccc' `cccc'.
cat bio-j-jsa-gut.wds \
| sed \
-e 's/^/_/g' \
-e 's/$/_/g' \
-e 's/[ql]j/H/' \
-e 's/[ql]g/P/' \
-e 's/cs/z/g' \
-e 's/ij/k/g' \
-e 's/ix/e/g' \
-e 's/is/r/g' \
-e 's/iiu/n/g' \
-e 's/y/i/g' \
-e 's/ci/a/g' \
-e 's/cg/8/g' \
| egrep '[^czHP]Pzcc[^czHP]' \
| wfreq
3 0.38 _qoPzcc8a_
2 0.25 _oPzcc8a_
1 0.12 _oPzcc8ae_
1 0.12 _oPzcc8_
1 0.12 _Pzccoe8a_
----- ---- ----
8 1.00 TOT
set noglob
foreach f ( \
'_qo'.'zcc8a_' \
'_o'.'zcc8a_' \
'_o'.'zcc8ae_' \
'_o'.'zcc8_' \
'_'.'zccoe8a_' \
)
echo " "
echo "-----------------------------------------------------------------------"
echo " "
cat bio-j-jsa-gut.wds \
| sed \
-e 's/^/_/g' \
-e 's/$/_/g' \
-e 's/[ql]j/H/' \
-e 's/[ql]g/P/' \
-e 's/cs/z/g' \
-e 's/ij/k/g' \
-e 's/ix/e/g' \
-e 's/is/r/g' \
-e 's/iiu/n/g' \
-e 's/y/i/g' \
-e 's/ci/a/g' \
-e 's/cg/8/g' \
| compare-contexts -rctx 0 -lctx 0 -colw 24 \
"${f:r}P${f:e}" "${f:r}[^P]${f:e}" "${f:r}[^P][^P]${f:e}"
end
unset noglob
-----------------------------------------------------------------------
3 1.00 _qoPzcc8a_ 7 0.88 _qoHzcc8a_ ----- ---- ----
----- ---- ---- 1 0.12 _qoezcc8a_ 0 1.00 TOT
3 1.00 TOT ----- ---- ----
8 1.00 TOT
-----------------------------------------------------------------------
2 1.00 _oPzcc8a_ 14 0.88 _oezcc8a_ 1 1.00 _oeHzcc8a_
----- ---- ---- 2 0.12 _oHzcc8a_ ----- ---- ----
2 1.00 TOT ----- ---- ---- 1 1.00 TOT
16 1.00 TOT
-----------------------------------------------------------------------
1 1.00 _oPzcc8ae_ ----- ---- ---- ----- ---- ----
----- ---- ---- 0 1.00 TOT 0 1.00 TOT
1 1.00 TOT
-----------------------------------------------------------------------
1 1.00 _oPzcc8_ 2 1.00 _oezcc8_ ----- ---- ----
----- ---- ---- ----- ---- ---- 0 1.00 TOT
1 1.00 TOT 2 1.00 TOT
-----------------------------------------------------------------------
1 1.00 _Pzccoe8a_ ----- ---- ---- ----- ---- ----
----- ---- ---- 0 1.00 TOT 0 1.00 TOT
1 1.00 TOT
Again the `P' seems to be similar to `H' and `e'.
And now for something completely different. Let's look at how the words are distributed among the
paragraphs:
cat bio-j-jsa.wds \
| sed \
-e 's/[ql]j/H/g' \
-e 's/[ql]g/P/g' \
-e 's/ij/k/g' \
-e 's/ix/e/g' \
-e 's/is/r/g' \
-e 's/iiu/n/g' \
-e 's/cy/a/g' \
-e 's/ci/a/g' \
-e 's/in/m/g' \
-e 's/ir/w/g' \
-e 's/cs/z/g' \
-e 's/cg/8/g' \
| enum-words-in-blocks -vWPB=100 \
| egrep -v '[^a-zA-Z0-9_ ]' \
| sort +1 -2 +0 -1n \
| make-word-location-map -vNBLOCKS=71 \
> .foo
The result has been posted as http://www.dcc.unicamp.br/~stolfi/voynich/word-distr-map.html
Recomputing with fewer large blocks:
cat bio-j-jsa.wds \
| sed \
-e 's/[ql]j/H/g' \
-e 's/[ql]g/P/g' \
-e 's/ij/k/g' \
-e 's/ix/e/g' \
-e 's/is/r/g' \
-e 's/iiu/n/g' \
-e 's/cy/a/g' \
-e 's/ci/a/g' \
-e 's/in/m/g' \
-e 's/ir/w/g' \
-e 's/cs/z/g' \
-e 's/cg/8/g' \
| enum-words-in-blocks -vWPB=1010 \
| egrep -v '[^a-zA-Z0-9_ ]' \
| sort +1 -2 +0 -1n \
| make-word-location-map -vCTWD=3 -vNBLOCKS=7 \
> .foo
cat .foo \
| gawk '/./ { printf"%5d %-16s ", $1, $2; for (i=3; i<=NF; i++) printf " %2d", int(($(i)*99/$1)+0.5); printf "\n" }' \
> .bar
Results posted in my Voynich page.