7.1 General Comments
The results of the study as discussed in section 6 speak clearly in a number of areas. In summary 90 these are:
Vocabulary Richness (6.3) - there is a general upwards trend over the text, and a large amount of internal variation between chapters.
Word Length (6.5) and Sentence Length (6.6) - these two also show an upward trend coupled with internal variation.
Soliloquy Length (6.7) - this shows an enormous variation over time, although without the clear patterns of the above two measures.
2.Language Use varies by Character:
Vocabulary Richness (6.3) - this shows very distinct differences between the characters.
Word Length (6.5) - this also shows large differences, with the exception of a local common peak in chapter IV, and two partial convergences in chapters VII and VIII.
Sentence Length (6.6) - again there are large differences, with convergences in chapters IV and VIII.
Soliloquy Length (6.7) - This measure provides less support, as most of the differences are in the magnitude of the change for each character, rather than in the direction.
3.Bernard's use of language is different from the other characters:
Vocabulary Richness (6.3) - this is much lower for Bernard than for any of the other characters.
Word Length (6.5) - this score is the highest of all the characters.
Sentence Length (6.6) - this is anomalously low for Bernard, but only for the body of the book. It is much higher in chapter IX.
Soliloquy Length (6.7) - also high during the body of the book, and highest overall because of chapter IX. The values for chapter IX in both Sentence length and Soliloquy length are at variance with his results for the other sections. This makes for quite a confused picture, very different from the other characters. Obviously, his position is necessarily unique by virtue of his being used to sum up the lives of the group. In this context the results for chapter IX are not so surprising, but the differences for some of the other measures are less readily explicable.
4.The language of the preludes is different in a consistent fashion from that of the characters:
6.Chapters IV and VIII are particularly important sections of the book for the characters:
As should have been obvious from the discussion of the critics' views in section 2, it would be impossible for all the critics to have been proven right. If we now compare these results with the views of the critics, what inferences can we make?
Assuming, as we have throughout, that language use reflects characterization, the evidence concerning variation in time supports the group of critics who argued for development of the characters. This was by far the largest group, consisting of Bazin, Brewster, Daiches, Gorsky, Kelley, Leaska, Moody, Rantavaara, Richter, and Thakur. In this case, the evidence from the computer supports the majority view.
The evidence for variation by character is also quite unequivocal. It provides good support for the group of critics who argued for clear differentiation, although there are some grey areas. As we have seen, there are indications of some sort of union of the characters in chapters IV and VIII. This evidence tends to support Gorsky's three-layered view of the characterization to some degree, although she maintains this obtains consistently throughout the book; The computer results do not suggest this. The results also provides support to the 'purist' group, consisting of just Daiches and Kelley. The evidence could just possibly be fitted into an interpretation which claimed that the characters are at the same time differentiated and unified, as many critics do, but there is no clear justification for such a view. The findings in this area thus go sharply against both of the two majority views on this area, the other of these being that the characters are either barely separable, or actually identical.
The evidence for the special position of Bernard among the characters lends itself less easily to ready linking with any one group of critics. Perhaps the most obvious association is with the group who believed that of all the characters, Bernard was the only one who had grown significantly by the end of the book. Unfortunately for them the evidence only shows Bernard as different from the other characters with respect to certain aspects of his use of language. It would seem unlikely that this should relate directly to his growth as a character any more than it should for the other characters. Change over time seems to be general for all the characters, although its precise manifestations may vary. Bernard's use of language in the final chapter is admittedly quite different from the rest of his speech, but not because he is the only character who has grown significantly. The function of chapter IX itself is quite different from the rest of the book (see section 7.3).
The results from all measures for the preludes come as no surprise; the critics are all agreed that the preludes show richer, denser, and more evocative use of language than the rest of the book. These quantifications of some of these aspects are however a valuable additional piece of evidence.
The importance of chapters IV and VIII also supports the view that the characters develop, in so far as it shows that the measures change over time. As far as the issue of the differentiation of the characters in these chapters is concerned, the evidence is less clear. While Susan is different from the other characters in use of pronouns in chapters IV and VIII, this difference is not reflected in the other measures. The conclusion must be that with respect to use of pronouns, length of words, and length of soliloquies in chapters IV and VIII, the characters are remarkably uniform. This could lend some support to the view that the characters are both many and one. In this case, as far as the evidence was concerned, one would have to say that the characters' use of some of the measures is similar in chapters IV and VIII, while remaining far from uniform for the rest of the book.
