INSTRUMENTAL  TRANSCOMMUNICATION
by Ernst Senkowski

   back to TOC 

B-12.2          STATISTICAL ANALYSIS

We are doing (hard) work so that we may really speak. 

Statistical analyses have furnished some remarkable information about the characteristic properties of tape-recorded voices (VOT). The most prominent quantifiable feature of the psychophonic style (discussed in chapter B-11.2 L2) is its much lower average ratio of syllables to words (S/W ratio) as compared to normal speech. Trajna compared 2 x 24000 words from normal texts of four experimenters with an equivalent number of words from voices on tape they had recorded. The reduction in the S/W (paranormal:normal) ratio was around 1,6 : 2,0, i.e.,  20 % [112]. The author’s analyzation of a smaller number of chance controls made on his own work, and that of others, produced comparable results (table Ill. 20). Although the values S, W, P, and their ratios show relatively great differences in the individual recording tests, the mean values S/P and W/P, i.e., the passage (or sentence) lengths increase with the time. The increase of duration linked with it indicates growing abilities in adaptation: the ‘synchronization’ of the different systems is maintained each time for a longer while. Parallely, S/W increases from 1,55 to 1,98 and reaches – on psychophonic style basis – the value of normal speech. The table does not reveal that, nor in which manner, the number of passages and their maximum lengths increased with every successive experiment. Looking on the whole of analyses, data flow and data flow rate have increased in the course of development.

One can avoid the time-consuming measuring of passage durations P [113] by limiting oneself to the determination of the number of passages P(n) with each n syllables per passage S/P and putting it over n. The thus resulting frequency distribution curves are linked by nature with the time distributions by the speed of speaking. The graphic of Ill. 21 shows each of the relative P(n)P percentiles. This illustration may be apt to induce some interesting reflections.

Ill. 20 :   SYLLABLES (S), WORDS (W), PASSAGES (P), AND THEIR RELATIONSHIP
IN VARIOUS TRANSAUDIO CONTACTS OF DIFFERENT EXPERIMENTERS

 Ill. 21:  TABLES AND GRAPHIC OF FREQUENCY DISTRIBUTIONS 

If one starts from the frequency distributions of VOT, the maximum determined by MACRAE with 1.50 sec. to 1.75 sec. corresponds to a range of 4-6 S/P. As the passages increase in length, the frequency distributions become flatter, broader, and less ‘smooth’. The usual linkage of: short sentences - low S/W ratio (and the reverse: long sentences - high S/W ratio) is not obligatory. Relatively long passages can be made up of short words: for example, Koenig, May 14, 1988: S/W = 1.72,  S/P = 15.5; or Haerting, 1987, selection: S/W = 1.77, S/P = 17.2. On the other hand, a one-word sentence can be made up of a very long word, e.g., Kontaktfeldverbesserung (contact field improvement) (CETL) has a S/W ratio of 7. Over the evolution of instrumental transcommunication, S/P values have risen considerably (see Figs. 20 and 21). Individual values vary over a wide range; the absolute maximum is a sentence with 55 W and 110 S, i.e. a S/W ratio of 2.00.  

MACRAE has shown how improbable it is that short meaningful passages would come about as a result of random ‘editing’  from a continuously spoken text. He is convinced that the maximum of the frequency distribution of VOT argues against their accidental origin. 

The interpretation of the paranormal sentence we must speak children’s language suggested in chapter A-2 can now be taken further: this kind of ‘speech’ is in some ways related to that of our young children [114]. A humorously serious computer text having arrived at HOMES from DEVAs (spirits of plants) shows the trait of young children’s way of speaking.

  back to TOP          September 2004