Privacy and Ethics
Within the context of creating large, publicly accessed archives that may include a variety of information resulting from the research process such as primary data, personal notes, correspondence, etc., one of the most pressing ethical issues that must be addressed is ensuring the privacy of the research subjects.
A summary of the major points arising during discussion are provided below. The reader should note that the group was in a better position to raise issues and consider the advantages and disadvantages associated with different solutions than to devise specific, concrete recommendations. We think this accurately reflects a lack of consensus at present regarding many key issues of ethics involving digital anthropological data.
1) Code of Ethics. Numerous societies have a code of ethics readily available via the web. See the web addresses below. Not surprisingly, these codes tend to provide general information or guidance regarding research protocols. Common themes include the following from the AAA guide “Anthropological researchers must do everything in their power to ensure that their research does not harm the safety, dignity (psychological well-being), or privacy of the people with whom they work, conduct research or perform other professional activities.” With regard to privacy it is suggested that “Anthropological researchers must determine in advance whether their hosts/providers of information wish to remain anonymous or receive recognition, and make every effort to comply with those wishes.”
2) Basic Privacy of the Individual. Common to all subfields was the basic privacy of the individual. If an individual wishes to remain anonymous then that wish must be honored.
However, ensuring anonymity is not a simple procedure. Common practices for deidentification of individuals include the use of a pseudonym or alphanumeric ID in place of the individual’s name. This will frequently not be sufficient for instance, in modern genetic analyses, it is relatively simple to identify individuals, families, and familial relationships from available data. Similarly, it may be a simple matter to identify an individual in linguistic or cultural work if the population of interest includes only a few individuals such as in the case of many endangered languages.
3) Protection of Location. In several instances, locations were identified as in need of protection. The most obvious of these may be in the case of archaeological site location where knowledge of the location may lead to unwanted access. Similarly, the location of sacred sites should be afforded the same privacy considerations as individuals or groups. Finally, knowledge of commercially exploitable resources within a populated area may need to be protected to prevent unwanted exploitation.
4) Length of Protection. While, at first thought, it may seem obvious that if an individual or group has requested anonymity, that request should be honored in perpetuity, this may not be the case. What may have seemed like sensitive information to a participant at one point in time may not seem so at a later point. The reasons for such changes in attitude may be numerous and variable. The dynamic nature of the concepts of “private” and “sensitive” must be kept in mind when setting up archives. In general IRB’s may suggest that the appropriate length of protection be “until no foreseeable harm can be done.” As noted this may be difficult to determine and may change as time passes. The decision to maintain or remove privacy criteria can reside both with a single individual as in the case of the donor or may be the responsibility of a group.
5) Restriction of access. One of the basic premises behind digitally archiving of data is to make these resources easily available to a group beyond that of the donor/collector. In many cases, the intention is to make the resources freely available to the general public without restriction. As noted, however, in many cases restrictions will have to be placed on access to protect privacy or to accede to a donor’s wishes. In short, it is clear that in some circumstances, differential access is sometimes a necessity. The most common differential access currently practiced is with regard to the dichotomy of scholars vs. nonscholars recognizing that the distinction between the two groups is sometimes difficult to identify (though some fields, like linguistics, also give special privileges to communities from which data are drawn. Restricted access is frequently placed on museum collections primarily for the safety of delicate collections but a number of online databases also require registration and validation of the user prior to access.In unusual situations, access may be restricted to a small number of individuals or even a single individual. Situations in cultural anthropology and linguistics were described where an informant had specifically stated that a given individual (e.g., a brother-in-law) could not be privy to the story being told, but that the rest of the world could be given full access. Such situations become very difficult to monitor.
6) Sensitive Data. In situations where access to restricted data is granted to specific individuals it is likely that certain covenants are placed upon the use. This may be as simple as the request that the user properly cite the donor/archive. It is also possible that stronger restrictions are placed on the use such as limiting the use of data for future studies or limiting the sharing of data between individuals. In terms of digital archives where sharing is easy, it is possible, and probably likely, that such covenants do not always transfer with the data allowing for violation of promised privacy by secondary or tertiary users. In these instances the question arises first as to how restrictions can be maintained and second as to how such violations can be addressed and violators punished. It was agreed that any action would be difficult and costly. For these reasons, it is likely that researchers may have to establish separate IRB protocol for archiving and web access for data in cases where it is decided to digitize and preserve the data after the initial research. It was also noted that some IRBs may mandate that sensitive data are destroyed following the project’s completion. If that is the case, the original protocol must be amended and an additional protocol submitted.
7) Privacy of the User. There was a significant discussion with regard to the rights of the user.
The discussion ranged from allowing the user total free access with an assurance of complete anonymity to a system requiring registration of users along with a login prior to access of the archive. Ease of access increases the potential use and benefit of the archive while posing no threat to the user. Restricted access (ranging from simple login to full-scale registration and authorization of users) may decrease the benefit to the user (if they decide to only focus on those archives with full, easy access) while increasing the benefit to the originator of the data. Benefit to the originator includes the ability to document usage which may be used in some way as a measure of the importance of the archive (potentially beneficial in situations such as promotion) or to provide assurance that archival material is used, and cited properly.
8) Spheres of Responsibility. Ultimately it was clear to the group that there are different spheres of responsibility for individuals at each level. The field researcher collecting data must first and foremost be responsible to the participants agreeing to maintain and protect the privacy desired by the individual. This is monitored by the IRB of the researcher’s institution. The archivist is likely to be removed from the original participants and is primarily responsive to the researcher. Just as the researcher has rules and restrictions placed upon them by the participants, the archivist can expect to have such covenants placed on them by the researcher. This again may be monitored by an IRB with ongoing protocols. Finally, the user, anonymous or not, has an obligation to treat the data with the respect that researcher and archivist have established. This will ensure the ethical treatment of all subjects and subject information.
LIST OF WEB RESOURCES REGARDING ETHICAL CONDUCT FOR THE PRIMARY ANTHROPOLOGICAL ASSOCIATIONS WITHIN THE USA.