Concerns about sensitive content in born-digital records seem to be a major factor in inhibiting the deposit of public records in dedicated digital repositories in Western countries. These concerns are much exacerbated by the changed nature of the process of reviewing records. The University of Glasgow, working in collaboration with the Foreign and Commonwealth Office, received funding to investigate the technology-assisted sensitivity reviewing of born-digital records. As part of this research, some preliminary research in a commonwealth country in Sub-Saharan Africa was carried out. The research, reported in this chapter, was carried out in Malawi by the late Dr. Mathews J. Phiri. He found that already there is a real, albeit limited, demand for technology-assisted sensitivity reviewing of born-digital records in Malawi. The available evidence suggests that within the next decade there is likely to be an increase in the need for effective means of assessing sensitivity in born-digital records.
TopIntroduction
For many years a spirited debate has taken place over the custody of digital records (Tough, 2004). There are those who argue that records created digitally should be retained by the creating agency and the future role of archivists should be as facilitators of access (Upward, 2000). In the opposing camp are those who argue that archives must continue to take records into custody, irrespective of the format in which they are created (Cunningham, 2011). In recent years, the second group has prevailed. Large and expensive digital repositories have been created in Australia, the United Kingdom, the United States and elsewhere. However, most of the digital repositories remain massively under-utilised. Short-lived public bodies like Royal Commissions and other enquiries do deposit their records as they are bringing their existence to an end. However, few ministries, departments or agencies do so.
Concerns about sensitive content in born-digital records seem to be a major factor in inhibiting deposit of public records in dedicated digital repositories. In Western countries, this concern is primarily of two types. Firstly, there is concern in relation to sensitive personal data of the kind that is covered by data protection laws (e.g. regarding a person’s denominational affiliation, sexual orientation, earnings, bank account details, et cetera). Secondly, there are concerns about national security, defence, anti-terrorist initiatives, et cetera). As Sir Alex Allan has expressed it, in relation to the British context (Allan, 2014):
The risk if this issue is not addressed satisfactorily is either that material will be released to TNA without proper review, leading to embarrassment when sensitive material is found to be in the public domain; or, perhaps as a reaction to the discovery of such releases, departments become risk averse and apply for blanket closures of records
Difficulties of the kind described by Sir Alex Allan are exemplified in events in the USA surrounding the release of e-mails created by Hillary Clinton as Secretary of State and controversially stored on a private server (Westwood, 2016). There were widespread complaints regarding the long delays in making these available: the delays were due to an acute anxiety that sensitive material might be accidentally made public. Approximately 30,000 e-mails had to be reviewed and, as the review was conducted by human beings without advanced technological aids, this proved a lengthy undertaking.
The concerns outlined above are much exacerbated by the changed nature of the process of reviewing records. Until quite recently, public records characteristically took the form of well-maintained filing systems, run by specialist records staff, which presented large numbers of individual documents in logical relationship to each other (Reed, 2005). This is no longer the case. Despite the availability of Electronic Document and Records Management Systems [EDRMS], the great majority of contemporary public sector records take the form of digital objects created and kept in poorly-organised systems that fail to create persistent links between related items (Moss, 2005). So reviewing often involves adopting a document by document rather than a file by file approach. This shift has tended to undermine long-established practices that involved retired civil servants, often people of high intellectual calibre, gathering to spend mornings reviewing files before going to an agreeable lunch. This productive human dynamic is not readily applied to the reviewing of born-digital records.