arnoud glaudemans, rienk jonker and frans smit documents, archives and hyperhistorical
societies: an interview with luciano floridi
functional way, this tends to be forgotten, and that would be a pity. We run the risk
that memory may be memorialised (mummified in some permanent, immutable
form) by the digital, instead of being kept alive as a trace of the past.
Then there is the whole debate about authenticity, which is of course complicated.
The authenticity of the digital has gone through a variety of technical solutions.
With the new and ongoing development of solutions like blockchain, the digital
seems to regain some credibility in terms of authenticity. In general, though, the
digital kills authenticity: there is no sense in asking what the authentic digital
document is and what an exact duplicate of it. A file and its copy are identical.
There might be a timestamp, but the timestamp is just metainformation, an
addition to the file. The two files themselves are interchangeable. They pass the
Leibniz test, so to speak: they are not distinguishable, and therefore they are
interchangeable. Leibniz used to talk about the identity of indiscernibles. If you
cannot discern the difference between A and B, A and B are the same thing. We are
used to that, since we started producing things in a repetitive way, from identical
vases to identical books. The digital has exacerbated a problem already caused very
visibly by the industrial. The more you produce indistinguishable things, like two
instances of the same iPhone, the more you lose authenticity as a concept that
applies to singular objects (call them tokens) instead of their model (call that type).
In this vein, the impact of the digital on authenticity is profound. However, with
new technologies, we are trying to find a solution 'with the digital for the digital'.
Blockchain is a solution, and there may be others. You can have a series of records
that keeps track of a particular file, giving a full history of who originated it, when it
was modified, by whom, and how.
This brings us back to the materiality of the digital. Because that blockchain 'lives'
somewhere. It lives off electricity, on computers that need energy. That is not trivial.
For instance, bitcoin is one of the currencies that use blockchain. And if massively
adopted, it would be an environmental disaster. To simplify, it would be like having
every coin in your pocket, every banknote you have in your wallet, behaving like an
electric bulb that needs to be kept switched on. Clearly, if we were to implement
bitcoin as a currency (like the Euro), this would consume a massive amount of
energy. That is the materiality we are talking about.
Now let us go back to the distinction I introduced before. The materiality of the
digital is of a different type than the materiality of the analogue. You cannot just say
that the archives are losing materiality through digitalisation. We are talking about a
different kind of materiality. So, we should better discuss the differences of the
materiality of the analogue world and the materiality of the digital world. Because
there is a difference. We need to figure out what care both kinds of materiality need.
To make a difference between material versus immaterial or nonmaterial, is highly
misleading.
EDITORS: There is also another aspect of persistence: that of the informational
content. Is it still useful to make it persistent? Information nowadays becomes more
fluid in how it works, functions, has meaning and is used. Should the term 'record'
be interpreted as a function to retain the persistency of the information contained
in it?
FLORIDI: Yes, that is a very good point. So, the persistence is not what qualifies an
archive. Of course, it has to be there, you have to go back to it, it has to be the same.
You might have grasped that my general strategy here is reshuffling existing
distinctions rather than endorsing them or abandoning them. In the case of the
digital, we have the problem of the fluidity. The digital rewrites itself and is easily
modified. How do we cope with the persistence of it? In the best scenario,
persistency is kept when a record is going through a series of modifications.
Consider for example Google docs. When you write a document, it keeps time
stamped copies, so that you can go back to previous versions. That document is not
persistent in the same way as a printed piece of paper is, it is malleable, and yet it has
a history of continuous changes that we may be able to reconstruct, contrary to the
single version of a paper document. This means that the digital can provide a
longitudinality of memory (all the several copies of a manuscript, for example)
much more easily than the analogue. This is very different from the recent past,
when all you could do was rewrite your file every time you saved it. The digital is in a
way keeping its 'persistence' by making sure that it keeps a record of the
sedimentation of versions. These are like snapshots of its development.
Unfortunately, most of the digital information that we have does not enjoy that
persistence, like for instance most websites.
The question is: what is the value of persistence? To me, it goes back to authenticity,
being able to say: that is what we agreed upon. You could say that it is truthfulness
that we are preserving in a document or in a record. Remember that writing was not
invented for postcards, or to tell people about our holidays. It was for legal reasons. It
was for keeping reliable answers to questions like: how many sheep do I have? how
many cows do I owe you? whose land is this? which rules have we agreed upon? I
think that that legal ability of records makes people coordinate their actions. It
seems to me that the value of persistency goes back to authenticity and therefore to
providing the reliability that something was "like this".
In the digital environment, we need to bind the record with the technologies
required to read it. Analogue records are not very demanding in terms of technology.
As long as there is a bit of light, as long as you can see, as long as you can read, then it
is okay. Today that is clearly not the case. A digital record without the right soft- and
hardware is as good as a magnetic pattern. In this sense, we are making the
materiality of records an issue that is even more serious than in the past. Because
now, not only do we have to take care of the material record, but we also have to take
care of the software and the hardware required to make it accessible. The problem is
so much present that it tends to disappear. It is difficult to explain to people because
for us materiality is what you bump into. But we also have the materiality supporting
the digital.
EDITORS: You could say that materiality has a lot of layers, which we have to take
into account. We account for the necessary layers and the rest we trust until
otherwise proven. It is all about trust, not truth. There was always a kind of trust in
the paper world. In the digital world it could work the same way. With some
processes it is necessary that there is for example a third party to authenticate, and
thus provide in this trust.
archives in liquid times
310
311