Analysis of word dependency

Dr. Vladimir Sazonov 18.03.2003

[Home]
General explanation

Here we analyze, how the Voynichese word depending on the previous word.
Only P,Q,R paragraphs was used, as having complete sentences.
The following groups are done because of statistical similarity:

K means any of k,t,p,f gallows
S means any of d,l,r,s dealers
O means o or a
M means m or g
< means "Begin of the sentence" (here it means a begin of paragraph or after "=")
> means "End of the sentence" (here it means an end of paragraph or before "=")

The statistics used for the previous word group w0 and the next word group w1 is

100*P(w1|w0)/P(w1)

The value "100" in this case means "no dependency", more as 100 means "often", less as 100 means "seldom".
The colour displays the level of dependency - strong read means "sufficiently more often as expected",
strong blue means "sufficiently more seldom as expected", weak red or weak blue means weak dependency.

First conclusions

1. The most spectacular is the Table 4, displaying very strong dependency between the final of the previous word and the initial of the next word.

It means "phonetical dependency" - that the next word changes itself dependent of a precious word.
The phonetical dependency is very often in arabic: for example the initial particle "AL" changes itself depending on a final of the previous word:

RAHMANU+AL+RAHIM --> RAHMAN-UR-RAHIM

2. As it is clear from the tables 1 and 3, the forms words very often reproduce the same form (the same initial or the same final) in the next word. For example, -aiin or qo- or -dy can repeat many times with good dencity.
It can mean (1) poetical text or (2) repeating the same morphology

Table 1.Initial-Initial dependency

Initials	<	K	O	y	S	q	ch	sh	e	cKh
36742	831	2866	9333	1692	6478	5338	5760	3175	110	870
<	0	967	36	36	33	22	5	8	0	66
K	52	100	97	90	98	86	99	142	58	117
O	138	72	130	112	88	77	99	92	135	60
y	67	92	99	159	104	84	103	86	118	117
S	116	63	95	126	111	79	105	123	97	115
q	31	46	94	70	105	132	119	114	43	53
ch	145	102	86	79	118	102	109	65	104	138
sh	69	123	77	75	70	201	68	109	136	118
e	80	93	146	59	61	87	92	105	607	76
cKh	101	61	78	139	146	71	87	90	76	300

Table 2.Initial-Final dependency

*	>	y	ey	dy	d	l	r	s	n	M	O
36742	831	4304	3827	6416	576	5708	5381	1106	5854	1038	1149
<	0	94	57	84	115	108	153	159	85	34	169
K	52	106	101	107	113	90	111	76	88	107	112
O	138	90	97	97	101	98	103	102	103	117	88
y	67	117	110	87	116	90	98	139	96	121	102
S	116	99	102	92	102	100	99	94	99	112	110
q	31	100	108	133	90	104	86	74	92	70	89
ch	145	101	93	84	98	101	99	107	112	106	99
sh	69	106	110	116	88	104	84	86	98	56	88
e	80	93	122	67	57	111	68	151	142	128	87
cKh	101	120	79	55	95	105	109	206	102	113	117

Table 3.Final-Final dependency

*	>	y	ey	dy	d	l	r	s	n	M	O
36742	831	4304	3827	6416	576	5708	5381	1106	5854	1038	1149
>	0	94	57	84	115	108	153	159	85	34	169
y	147	130	89	87	106	96	93	102	102	84	106
ey	83	87	166	89	80	97	79	91	113	62	85
dy	100	82	97	165	119	88	80	71	86	84	69
d	69	103	85	109	177	93	111	115	88	49	138
l	54	85	86	113	99	118	99	108	97	105	85
r	68	107	90	75	101	98	127	91	102	144	82
s	83	94	92	45	74	94	114	177	139	108	135
n	130	122	106	82	79	104	96	88	97	108	98
m	353	50	78	63	98	83	111	163	114	180	172
o	50	92	89	63	105	104	114	135	111	61	253

Table 4. Final-Initial dependency

*	<	K	O	y	S	q	ch	sh	e	cKh
36742	831	2866	9333	1692	6478	5338	5760	3175	110	870
>	0	967	36	36	33	22	5	8	0	66
y	147	105	74	113	143	104	85	80	77	109
ey	83	142	72	57	131	176	62	50	61	102
dy	100	74	77	77	100	219	74	63	83	44
d	69	42	103	177	105	139	73	88	115	36
l	54	119	84	74	120	57	128	137	157	88
r	68	29	163	107	49	37	125	138	148	111
s	83	27	192	137	46	32	94	99	151	99
n	130	22	124	120	62	55	143	141	62	148
M	353	91	75	290	147	59	58	79	64	44
O	50	176	47	64	223	60	62	50	29	246