Under Construction @ Keele 2016 Volume 2 Issue 1 | Page 34

26

them, provided a wonderful opportunity to uncover a previously hidden
history of semi-penal institutionalised women.45
The historical data was analysed in a similar manner to that employed to analyse
interview data. However, rather than coding the material in the traditional sense, I
employed a deconstruction and re-assessment of the records, reports, letters and
minutes within the context of a Foucauldian feminist framework.

The resultant

reconstruction of past events was based upon ‘an intelligent selection and
interpretation of the best evidence which is available’.46 William H. McDowell points
out nonetheless, that it is often difficult to read an historical document ‘without
consciously seeking to determine whether it was intended to manipulate its intended
audience or merely used as a means of self-expression’.47
Limitations of Historical Documentary Research
It was clear early on in the research process that a large number of the documents
and sources contained within the archives would be incomplete and fragmented,
difficult to decipher and very brittle to handle. Additionally, due to the relatively
limited quantity of archival evidence, such an exercise was more complex and timeconsuming than first anticipated. Although secondary analysis is typically considered
less time-consuming than performing a new study, this research has been inten