5 Conclusion
The factors governing the choice of be or have with past participles of intransitive verbs are prescriptive grammatical rules, linguistic parameters and extralinguistic factors.
Earlier research on the paradigm has isolated some linguistic parameters which have been found to influence the choice of variant form. In this study, the factors which proved most influential in the direction of have were the conditional and the use of modal auxiliaries. The most be-promoting features were adverbs expressing recent action or result and the -ing form.
Some verbs are difficult to classify in terms of transitivity/ intransitivity and may also be unclear as to their verbal or adjectival status. Including such verbs in an investigation concerned with participles of intransitive verbs may be harmful to the statistics.
Of the verbs found more than 10 times in the material, the clearly actional verb pass was the only one to have a have-percentage of more than 50%. Two of the verbs in this group, namely the resultative verbs grow and become, were not found with have.
The small time span of the corpus and the relatively small number of examples did not make an in-depth diachronic study possible. The paradigm seemed however to be slowly developing during the period in the direction of a more have-friendly average.
Among the extralinguistic factors possibly influencing the choice of variant form in the paradigm are the education and status of the language user. The results of the investigation show that the paradigm during the period examined may have been developing "from the top", that is, that the increased use of have might be the result of a fashion among those of higher rank and education rather than a change among the uneducated people.
A more uncertain conclusion is that spoken language may have been more conservative than written language. Neither the comedies nor the diaries are true representatives of spoken language. Still, it is possible that the difference between these text types and the letters (which undeniably are representatives of "true" written language) in terms of frequency of have+PP has to do with the difference between spoken and written language.
Further study of the be/have paradigm could include an investigation of the sociolinguistic factors which might have influenced the paradigm, to see if the shift from be to have indeed started among the educated classes rather than the people of lower rank, as indicated by this study.
Another topic, the thorough investigation of which would facilitate further studies of the type presented in this paper, is the verbal/adjectival status of certain PPs like done, passed/past, resolved and agreed.
Finally, this study has shown that a corpus composed of different text types is an aid when making a synchronic study in diachrony, since it augments our chances of finding evidence from different types of language of a period.