arnoud glaudemans, rienk jonker and frans smit beyond the traditional bounderies of
archival theory: an interview with eric ketelaar
Michetti rightly points out that the principle of provenance is crucial, also in the
digital context, in preservation, use, access, appraisal, arrangement, and
description. That is important, because mostly provenance is only taken into
account when arranging archives. Clearly it is also crucial in other fields and
processes. The more so when it is linked to the concept of context.
Van Otterlo explores the ethics of digital archives with an emphasis on the role of
algorithms. He describes a shift from 'the intended archivist' towards 'intentional
algorithmic versions' that could be linked to other essays. What I do not like is his
stressing the gate-keeping role of archivists. As a view of what archivists do, it is an
incomplete one, to say the least.
EDITORS: What is your opinion about the idea Van Otterlo's seems to offer, that you
could operationalise a code of ethics in a ruling for designing algorithms? This idea
of 'automating ethics' could be relevant not only for access, but also for (other)
processes of recordkeeping. What would you say about this idea of 'automating
ethics'?
ERIC KETELAAR: I would argue that both an ethical position, and an automated
facility regulating access, are politically framed. When, for instance, I subscribe
ICA's Universal declaration on access, it is a political decision. There is no natural
law stating that archives should be accessible to anyone. That is something that is
only 200 years old. What most archivists do not realise is that availability,
accountability, findability, etc., are not universal and natural laws or principles. It is
the law, yes, but the law is only an expression of what society at a particular point in
time believes to be right or wrong. Different ethical positions are possible, and
choosing one is a political decision. The outcome is therefore constructed, based on
a particular world view. The same counts when I automate access, when I design a
digital decision-making process to regulate access. I am talking about 'politics' with
a small 'p', of course. This is the nice thing about fluidity, about postmodernism.
Postmodernism, as Lyotard says, is: not believing the grand narratives, and asking
this sort of questions: why is availability necessary? Has that been put to a test? I see
a congruence of the political constructedness of, on the one hand the human,
ethical decisions about access, and on the other hand, any automated system
providing access. The algorithmic procedures have to be designed, and anything that
is designed, is designed within a political, ethical, etc. framework. And of course,
there is the fear of human behaviour being guided through 'hidden' algorithmic
procedures. But you should not forget that human decisions are, in a way, hidden as
well.
EDITORS: Algorithms seem to have something mystical about them: they are
hidden, so you have to fear them. Could we conclude that you would say that the
bias in algorithms - hidden or non-hidden - in the creation of information is an
aspect of context and not of the 'record itself'? How would you 'situate' this?
ERIC KETELAAR: Well, coming back to what we discussed earlier: if indeed the
(informational) meaning of a record is not in the record itself, but in the eye of the
beholder or user, you could conclude that in an automated system or algorithmic
procedure, the meaning of a record is created by 'the eye' of the algorithm. It would
be a matter of accountability whether we can oppose such a meaning as given by the
algorithm; can we bring the algorithm to court, so to speak? Who is ultimately
accountable? Accountability is the essential thing here.
EDITORS: When interacting with a human being there is something the Germans
call 'freies Ermessen', which is not possible with algorithms. With a human being
you can interact in the sense of interference, discussion, and influence.
ERIC KETELAAR: Algorithms are programmed, and usually they are not 'fuzzy'.
Compare also the discussion about robotics and artificial intelligence.
EDITORS: To conclude this interview we would like to ask a last question. Given our
doubts and uncertainties concerning the archival theory and profession today:
Where, do you think, the profession will stand in, let's say, fifteen or twenty years?
ERIC KETELAAR: I anticipate the development going steady but more slowly than
most people think, are afraid of, or hope. The big challenge is that the show must go
on, the shop is open, and business has to be continued, while at the same time we
have to adjust theory, methodology and practice, and think about what these other
disciplines have to tell us. I am concerned that archival professionals will not take
time for really sitting down (or standing up) and have the sort of discussions we are
having right now. It is important that, in the Netherlands, we continue the
professionalisation in the way we started when the training of archivists was moved
from the School of Archives (Archiefschool) to the university. We now have the
opportunity of a more theoretical treatment of archivistics than we had in the past.
The same is true for other countries. We should be more open and hospitable to
other disciplines, as for example in the United States, with its Archival Education
and Research Initiative (AERI). In the Netherlands, as things have developed until
now, we still tend to be very much 'inward looking' into our own discipline. We have
to be more open to other disciplines. That is pretty much the case already at the
university, but it would also be important for practicing archivists and their
professional bodies. They should take the time and the opportunity to partake in
theoretical and methodological discussions like the one we are having right now. It
is of great importance that theory and practice stay closely connected, and we
should find structures for that, like reading groups, panels, blogs etc.
I believe that in the next fifteen years archival theory will not change drastically, but
will adapt itself to connect with changes in society. I partly base this, my optimism,
on the fact that in your book basic concepts of archivistics are being discussed and
adapted to a new environment. So, in a way, this collection of essays is a proof that
archivistics can move through time: adapting, inhaling, and infusing from other
disciplines. Therefore, let us try to keep up and maintain the triangle of theory,
methodology and practice. All these aspects should stay closely connected. I am also
curious what the increase in foreign students studying archivistics at the university
in the Netherlands will bring us, enriching archivistics in this country, and abroad.
So, there are a lot of signs of archivistics being, and staying, a truly vibrant discipline.
archives in liquid times
304
305