able to authenticate itself. Nobody else can do it. The authenticator must be inside the authenticated. Human beings should help the authenticator for doing its job right. We should enable the hyperobject to authenticate itself. This is one step further than Authentication by Design. Authentication of a hyperobject presupposes artificial intelligence that will enable records to learn to authenticate themselves. There is no outside interference in the authentication process anymore. Concluding remarks The question that this essay tries to answer is: can this confusing word, authenticity, be an applicable qualification for records in a digital environment? To answer this question, the following sub-questions were raised: 1. How is the term authenticity used in respect to records? 2. Does the nature of records change in a digital environment? 3. Can authenticity still be used as a qualification even when the nature of records changes in a digital environment? 4. Can we speak of a paradigm shift regarding the term authenticity in a digital environment? At this point, the answers to these questions are in my opinion: 1. It is used in a strong meaning, being a qualification of the nature of a record, and in a weak meaning, being a qualification of the way in which the record confirms the identity of another entity, for example a person, a group, an object or an event; 2. Yes, they have become the results of data processing; 3. Yes, even though authentication of digital records (when considered part of a hyperobject) is a process that might need new technologies; 4. No, see below. In my view, authenticity therefore remains an applicable qualifier for records. However brief and sometimes speculative this essay is, my conclusion is that, even if we see digital records as part of a hyperobject on a totally new scale, both strong and weak authenticity continue to be of vital importance. Authenticity remains a very sticky word. In my view it will continue to play a crucial role, despite all complications, paradoxes and obscurities that some with it. As to the last sub-question the answer is that there is no paradigm shift. There is no incommensurability between using the word in the context of analog records and in the context of digital records. It seems that the enormous acceleration (or growth) of records made possible by the changed nature of records in a digital environment, is the level on which a revolution is taking place (Sloterdijk, 2016, p. 15-20). It is like we are witnessing the emergence of an uncontrollable Open Archival Information System (OAIS) running on steroids. That would mean that the incommensurability of words does not take place on a fundamental conceptual level, but on logical, methodological and instrumental levels. I would like to conclude this essay with some closing remarks. Authentication is a highly ethical process. The verdict of authenticity is based on values of reliability and trustworthiness, either of the record itself or of the reference made to the record. Technological issues will be predominant in authentication processes concerning the nature of records. How far can technology go to be able to Authenticate by Design and which parts of technology should be affected by it? Will it affect the algorithms, the integrity of the data and the bitstreams? And will it be possible to create artificial intelligence for records, so they can learn how to authenticate themselves? Issues on power and awareness will be predominant in all authentication processes. Who owns the future in terms of strong authenticity of records? And who owns the future concerning the distribution of records seen from the perspective of weak authenticity? To explore the ethical challenges Floridi's ethical RPT model can be useful (Floridi, 2010, p. 104). This model gives the opportunity to define ethical actions of informational agents towards digital records. It makes a distinction of three types of actions: using information as Resource, Product and Target. Especially in producing (or: creating) and targeting (or: management) of digital records there is lot of research to be done to enable the authenticator. Both the hyperobject and its qualifications of strong and weak authenticity have the tendency to withdraw after becoming manifest. They play a game of assessing, appearing and disappearing. The game is highly relevant in present day discussions about filter bubbles, fake news and accessibility to evidence. Records are not about whether the information is true. However, the claim of truth can only be made by referring to sources (including records), that have a strong authenticity. In this essay I have avoided to use the word "archive", to concentrate on its constituent parts: records. The reason is that there are many definitions of an "archive" for example: the archive as the physical place where records are kept, the archive as the recordkeeping institute, the archive as the collection of records that constitute the representation of an organizations or the archive as the collection of records that represent the memory of a city of a country. Each definition would require its own considerations regarding authenticity in a digital environment and regarding high level concepts as hyperobjects (Morton, 2013) and infosphere (Floridi, 2010). The nature of records has changed dramatically on different levels of abstraction: being transformed from locked-up, scarce information, that has survived natural and man-made destruction, sticked to material and put away in specific locations, into innumerable objects of information that fit well into the five properties Morton describes for hyperobjects. The image of digital records as being part of a hyperobject consisting of digital information fits well into Floridi's concept of the fourth revolution as a "blow to anthropocentrism" (Morton, 2013, p. 36). archives in liquid times 262 frans smit records, hyperobjects and authenticity 263

Periodiekviewer Koninklijke Vereniging van Archivarissen

Jaarboeken Stichting Archiefpublicaties | 2017 | | pagina 133