Wednesday, December 28, 2011

AAA Draft Code: "Easily Remembered" or Overly Simplistic?

The American Anthropological Association (AAA) has released a draft code of ethics, the latest step in a revision process that began in late 2008, as well as a Final Report of The Task Force for Comprehensive Ethics Review. An Executive Board subcommittee is taking comments until January 30, 2012, at

As a non-anthropologist who respects disciplinary differences, I don't mean to tell anthropologists what to do, and I do not plan to submit a comment to the subcommittee. But I can point out that while the two documents represent an impressive effort, the draft code does not reflect all the concerns of some anthropologists who have thought seriously about ethical obligations.

The draft code lists "core principles . . . expressed as concise statements which can be easily remembered for use by anthropologists in their everyday professional lives."

The principles are

1. Do no harm
2. Be open and honest regarding your work
3. Obtain informed consent and necessary permissions
4. Balance competing ethical obligations due collaborators and affected parties
5. Make your results accessible
6. Protect and preserve your records
7. Maintain respectful and ethical professional relationships

Let's take a look at them.

Do No Harm

As I noted last year, when this principle was first proposed several scholars raised objections while others supported it.

The Task Force report acknowledges this debate:

Past codes specifically stated that anthropologists owe their primary ethical obligation to the people they study. While the members of the Task Force were sympathetic to this view, it became increasingly clear that it reflected a cherished anthropological value rather than an actual principle of ethical practice. Anthropologists "studying up," studying those in power, do not owe a greater ethical obligation to powerful individuals than to those vulnerable to that power. Nor is that value equally applicable to all kinds of anthropologists without either broad exclusions or special pleading (e.g., archaeologists, paleoanthropologists). While acknowledging the problematic nature of this previous principle, the Task Force nevertheless did discuss concerns that its removal weakens what had traditionally been perceived to be clear guidance for anthropologists caught in conflicting positions between the needs of research participants, sponsors, and other populations. We wish to note that this was a difficult issue for the Task Force, and we were never able to reach unanimity.

The new code does place responsibility for one's actions squarely on the anthropologist, however. It requires her/him to consider the impacts of the work and its potential to cause harm. The Task Force is keenly aware that the Do No Harm principle is also complex and problematic, yet we feel this more directly and immediately addresses the ethical imperative informing the older "primary responsibility" statement, while recognizing that anthropologists study all kinds of individuals and institutions, some of whom do not necessarily command our primary allegiance or obligation. As was the case with the 1998 revisions to the Code, this topic demanded much energy and emotion. We think the intense investment in this discussion – as we witnessed it among ourselves and in the wider membership through comments left on the blog -- is a strength of our Association, and we assume attention to and investment in this topic will continue among our members for the foreseeable future.

I see no such complexity in the draft code itself. Can an anthropologist studying "those in power" ethically expose their wrongdoing? The report seems to say yes, but the draft code says no.

Nor do I see that draft code's blunt language as consistent with the AAA's response to the regulatory proposals of this summer:

The research world is rife with projects whose results bore, interest, annoy, please, anger, or enlighten research participants (just as they do fellow researchers). This world is not generically the people's enemy. On the contrary, those of us working in US colleges, universities, news media, and research institutions have inherited traditions of free inquiry whose continuation is vital to this country's political, economic and social life. It would be deeply ironic if a regulatory system put in place to protect human beings were transformed into a device focused on restricting their power to know the world.

The draft code's explanatory section on "Do No Harm" does note that "Anthropologists may choose to move beyond research to a position of advocacy." But I don't think advocacy and critical inquiry are the same thing. One can enter and leave a research project without becoming an advocate, yet still bore, interest, annoy, please, anger, or enlighten research participants.

Be open and honest regarding your work

The clear intent here is to discourage, in the task force report's terms, "clandestine research in which informed consent could not possibly, by design or context, be adequately and fairly given." In other words, no spying for the CIA.

It is not clear if the code considers the duty to be open and honest regarding one's work to extend to, for example, oppressive governments. As Brian du Toit wrote in 1980,

There are persons who feel that certain kinds of research have to be conducted in a clandestine manner, and that unethical behavior is somehow excusable. Such research might involve subversive groups, the American Communist Party, the Ku Klux Klan, and depending on your philosophical persuasions, even the Pentagon. Pierre van den Berghe says of his research in South Africa that "from the outset I decided that I should have no scruples in deceiving the government. . ." Does informed consent become arguable when a social scientist studies groups or organizations who are by his/her evaluations dishonorable, anti-social, or morally outrageous? Should a researcher be ethical when studying superordinatates? Through whose definition are these moral evaluations reached? [Brian M. du Toit,"Ethics, Informed Consent, and Fieldwork," Journal of Anthropological Research 36 (Autumn 1980): 274-286, JSTOR.]

Du Toit concluded that "the social scientist is obliged to represent the true nature of his study, to clarify aims and methodology to those to be studied, to treat them and the information gathered with dignity, and to produce protection or support promised to those studied." I think this means that he considered van den Berghe's deception of the South African government unethical, though it's not clear from the essay.

In any case, I cannot find in either the draft code or the task force report any guidance on the question of deceiving authoritarian governments or subversive groups.

Obtain informed consent and necessary permissions

The draft code is somewhat nuanced, explaning, for example, that

The informed consent process is necessarily dynamic, continuous and reflexive. Anthropologists should initiate this process as a part of project design and continue through implementation as an ongoing dialogue and negotiation with research participants. Informed consent does not necessarily imply or require a particular written or signed form. It is the quality of the consent, not its format, which is relevant.

In 1996, Murray Wax disparaged the "dogma of informed consent" on the grounds that

"Informed consent is meaningful when a specific procedure is then to be administered at a specific time to a specific (powerless) research subject. However, in much ethnographic research, the researcher is not about to administer a specific procedure, but is initiating a long-term and open-ended social process among a group which typically possesses major powers in its own right. Because the social process will be jointly constructed by the investigator and the numerous and varied members of the host community, its outcomes are unknown, although the fieldworker's desired goal is recognition as a fellow human being and so allowing a form of membership and participation. [Murray L, Wax, "Reply to Herrera," Human Organization 55 (1996): 238]

The AAA draft code defines "informed consent" to resemble more the process described by Wax than the terms of the Belmont Report. But in doing so, I wonder if it risks confusion when anthropologists and IRBs find themselves using one term with two different meanings, as may also prove the case with "do no harm."

Moreover, the code fails to address some of the questions posed by du Toit back in 1980. Du Toit posed several questions about consent by children; when can a parent speak for a child when deciding to offer an object or photograph to a researcher? At what age can a child consent? As children reach maturity, can they change their minds about a study?

I'm guessing I know the answer here: it depends on the context. As the AAA response to the ANPRM so nicely put it,

even reference to "children" and the like only make sense in light of particular conditions and local cultural conditions. The Common Rule specification of "vulnerable populations" may make sense in most situations in the US, but may not make sense in other settings where social categories may differ (e.g., where the relevant distinction may be ritually initiated vs. uninitiated boys).

I am less certain what the draft code would suggest about the duty of getting consent from superordinates. Du Toit asked, "If a person's behavior is public, such as that of an office manager, company director, or police chief, but this person refuses consent for a study of that office, is the researcher at liberty to expose that person by name? Is the researcher exposing the public role, or also entering into the private realm of that official?" I see no answers in the draft code.

Balance competing ethical obligations due collaborators and affected parties

The gloss on this principle reads in part, that

Anthropologists have an obligation to distinguish the different kinds of interdependencies and collaborations their work involves, and to consider the real and potential ethical dimensions of these diverse and sometimes contradictory relationships, which may be different in character or change over time. When conflicts between ethical standards or expectations arise, anthropologists need to make explicit their ethical obligations, and develop an ethical approach in consultation with those concerned. Anthropologists must balance these competing ethical obligations while recognizing their obligation to avoid harm to those they study. Anthropologists should not agree to conditions which inappropriately change the purpose, focus, or intended outcomes of their research, nor should they mislead sponsors or collaborators about the nature of the work or its outcomes. Anthropologists remain individually responsible for making ethical decisions.

This passage is particularly hard to to understand without context or examples.

I was struck by the acknowledgment that anthropologists may need to recognize "differing ethical frameworks of collaborators representing other disciplines or areas of practice." I think this means they should not be surprised when other scholars fail to share their outrage about cooperation with the U.S. military.

Make your results accessible

This one seems relatively clear. While less absolute than the "do no harm" principle, it sets up the sharing of research findings--especially with research participants--as the norm, but admits exceptions "such as where participants have been fully informed and have freely agreed to limited dissemination."

This section notes that "Proprietary, classified or other research with limited distribution raises complex ethical questions." It's not clear if this includes publication in professional journals that are open only to paying subscribers and institutions. The AAA has shown leadership on this front, making its publications available free or at reduced prices to underresourced institutions around the world. But neither the code or the task force report mention such initiatives explicitly, so I just don't know how far this principle goes.

Protect and preserve your records

Like the previous principle, this one states a norm ("the interests of preservation ordinarily outweigh the potential benefits of destroying materials for preserving confidentiality") but also points to legitimate exceptions to that norm. It takes a similar approach to the ownership of the anthropologist's records. Normally, they belong to the anthropologist; but other arrangements may be appropriate.

Maintain respectful and ethical professional relationships

This section applies not to human subjects research per se, but to professional obligations not to "plagiarize, nor fabricate or falsify evidence," and to acknowledge the contributions of others, including students.

Final Thoughts

In sum, the draft code strikes me as uneven. Some sections appropriately acknowledge that rules have exceptions, but the earlier passages fail to allow for the possibility that scholars have the right, and even the duty, to hold people accountable for their public behavior, and to evade unjust restrictions on liberty.

The code also leaves me pessimistic about the value of codes in general. A set of principles followed by some paragraphs of explanation for each is better than just a bare list. But I am inclined to agree with Patty Gray's comment that

the circumstances of anthropological research done for corporate or marketing purposes is going to involve a slightly different set of ethical issues than anthropological research done for the purpose of discovering new knowledge about human beings and their lived experience, and sharing that knowledge in venues such as teaching, scholarly publications and academic conferences, popular communication (non-academic writing, public lectures, etc.). If you try to mix together these rather different "flavours", you are just going to get a mess that no one can swallow. I would rather see you serve them all up with their own individual integrity and let their nuances show through.

I am therefore more excited by the task force recommendation that

The AAA should produce and periodically update a publication of case studies of ethical dilemmas anthropological researchers, teachers and practitioners might face, suitable for use in graduate training, post-doctorate training, and continuing education.

This was the approach taken by the American Psychological Association in its 1973 Ethical Principles in the Conduct of Research with Human Participants, and by the AAA in its 1987 Handbook on Ethical Issues in Anthropology. (Not mentioned in the task force report; I wonder if they knew about it.) More recently, the Macquarie University ethics training module has taken this case-study approach.

These documents are easier to understand, and more likely to provoke reflection, than the abstractions in the code. Moreover, since the code clearly responds to specific controversies (notably over Human Terrain Systems) without naming them, a document presenting case studies would be more transparent than the ones here. I would go so far as to say that a collection of case studies should have preceded the development of more abstract guidelines, so that everyone would know what was at stake.

Note 1. In addition to stating the principles, the draft code notes that "The American Anthropological Association does not adjudicate assertions of unethical behavior, and these principles are intended to foster discussion, guide anthropologists in making responsible decisions, and educate." That is not to say that it will never again seek to do so; the task force recommends "that if the Executive Board wishes to pursue an adjudicative code as a possibility, it should appoint a committee to consider this matter only after the EB has determined if it is in a position to make a financial and philosophical commitment to this process."

Note 2. The AAA has a lousy server. This blog post was delayed by multiple periods in which I could not access the two main documents and their supporting materials. This problem was aggravated by the decision to post each section's bibliography as a separate PDF file. Make your results accessible, indeed!


Anonymous said...

Daniel Lende points out that the AAA is being less than consistent in executing the principle of making results accessible with their recent stand on public access to scholarly work.

Zachary M. Schrag said...

Thanks for this comment.

I share some of Mr. Davis's fears that open access could wreck valuable editorial systems. But I also share your puzzlement at his failure to mention the AAA's principled commitment to the dissemination of knowledge or that commitment's manifestation: the AnthroSource Philanthropic Initiative.