Creating value of the past through negotiations in the present: balancing professional authority with influence of participants

This article investigates how an online forum for a participatory archive project can offer a platform for controlled power-sharing between archivists and participants. Additionally, it explains how participants affect the processes, structures, and end usability of the information resource they are invited to create, with a focus on representability and the ethical responsibility of archivists. This study takes its point of departure in a participatory online transcription project: Begravelser 1861–1912. The influence of participants is studied through observation of the communication between archivists and transcribers in an online discussion forum. Theoretical concepts of maximalist and minimalist participation are used to contextualize levels of user influence. The study uses a framework inspired by Community-based participatory research (CBPR) to analyze the communication. The results show how participants used the forum to gain influence in the project, and how several elements of CBPR—shared influence, mutual development, and mutual use—were manifested in the forum. The conclusion is that using a CBPR approach in participatory projects can produce synergetic effects of increased knowledge and enhanced archival responsibility in conjunction with sustainable participatory engagement.


Introduction
'Participation is ultimately about power sharing, and if this is structurally absent or systematically undermined, then whatever is being called participation must be seen with the utmost skepticism, or indeed labeled fraudulent' (Dahlgren 2013, pp. 28-29).
A benevolent approach to user participation in archives, by way of various 'participatory archives' projects, is often framed as a way to counteract selectivity and bias in archival collections, archival descriptions, and finding aids (Light and Hyry 2002;Anderson and Allen 2009;Farley 2014;Eveleigh 2014). Achieving more representative collections, it is further argued, implies 'radical participation' (Duff and Harris 2002, p. 39), with changed dynamics in central archival activities such as arranging, describing, and providing access to collections (Gauld 2017). This requires modern archivists to be collaborative and 'open to user needs' (Baxter 2011, p. 299). The social role of archives and archivists has clearly changed.
On the assumption that archives are the mirrors of the society in which they exist, user participation is advocated as an ethically responsible way of constructing and administering archival collections (Gauld 2017;Brilmyer 2018). Inviting people with different backgrounds, interests, and perspectives to take part in archival processes enables them to make-through their varied knowledge and experience-an essential contribution. It helps make collections more representative of all aspects of society-in keeping with the mission of the archivist, which is to ensure that collections mirror society in all its complexity (Cook 2007;Schwartz and Cook 2002). However, unfiltered inclusion risks conflicting with the idea of archival custody of records, and the professional authority which that task entails [a clash discussed by Phillips (2014) and Yakel (2011), among others]. Since antiquity, archivists have been expected to safeguard the integrity and authenticity of records. In ancient Athens, any man found to have altered documents preserved in the city's central archive could even be sentenced to death (Duranti and Franks 2015). The image of the archivist as gatekeeper still holds true, in regard both to collections and to other information resources, the content of which must not be inaccurate, altered, or of false provenance.
In short, the participatory turn presents challenges to the archival profession and its working methods. It requires solutions for balancing participatory influence with professional authority, so that archival custody can be guaranteed simultaneously with radical participation. On these premises, this article investigates if and how an online forum for a participatory archive project can offer a platform for power-sharing between archivists and participants. Additionally, it explains how participants affect the structures and end usability of the information resource they are invited to create, with a focus on representability and the ethical responsibility of archivists.

Constructing a database from records of the dead
The online discussion forum that is the subject of this study is part of Begravelser 1861-1912, a project run by Copenhagen City Archive that aims to construct a database of information from funeral records. The records in question, which register every funeral that took place in the city, contain information categories such as name, address, occupation, and cause of death ( Fig. 1). Historically, those data were used for graveyard administration, as well as to keep track of contagious diseases ravaging the city. In the twenty-first century, the records are popular material for amateur genealogists. Their popularity, combined with their potential as research material, make the records an excellent source for a participatory transcription project, as has their accessibility, findability, and suitability for conversion into digital form for quantitative processing. Due to the requirements of structuration and enabling of machine readability, Begravelser 1861-1912 is an example of structured transcription. As opposed to other approaches to transcription-such as diplomatic transcription, where the records are transcribed faithfully down to every letterstructured transcription aims to transcribe only relevant pieces of text into already predefined database categories (Ridge 2014, p. 6). By November 2017, when the project had run for a year, participants had managed to transcribe the records of 105,000 funerals (#3968). 1 One hundred and three participants had been active in the project, and 25 of the most active participants had completed 96% of the transcriptions (100.000 begravelser tastet!, 2017). According to the Copenhagen City Archive, participants describe themselves as amateur genealogists, or as retired people with experience from occupations where it was important to have a good knowledge of medical terms or an ability to read handwriting (e.g., having worked in the library sector or as a medical secretary). However, persons with an entirely different background also participate in the effort (Stoktoft Overgaard 2017).
The transcriptions are performed in an online tool called Kildetasteren. 2 Linked to the tool is a user forum (to be found at www.kbhar kiv.dk/forum ), where participants can communicate both with their fellow users and with archivists ( Fig. 2). Together, the transcription tool and the forum offer a platform for participation and for communication between participants and archivists. The forum provides a space for participants to chat about observations, get help with blurred handwriting, or ask about strange causes of death. In a sub-topic of the forum called 'Feedback on Kildetasteren', participants can suggest improvements and modifications to Kildetasteren and to the funeral database. Improvements may consist, among other things, of changes to the use of existing fields, of modifications to the relations between fields, and of new categories of information from the records to be included in the database. Although the project is profiled as a participatory project-where the archivists put a high priority on ensuring 'that participants are involved and are listened  '(Interview,personal communication,16 November 2017)-archivists retain the right to decide which changes are implemented and which are not.
Decisions on specific suggestions depend on several other factors as well, like economic resources and the needs and requirements of third parties, such as academic researchers. Each update of the structure of the database, or of the information it contains, requires a careful balancing of these various factors. By reviewing suggested modifications and observing whose interests are reflected in the changes made in the database, we can gain insight into how the influence of participants is balanced with the professional responsibility of archivists.

Models of influence in participatory archives
Even though several successful examples of participatory transcription have been documented, the issue of volunteers' influence on a structural and transformational level is rarely brought up. It is not unusual to see statements emphasizing how crucial 'open communication with participants' is for building trust and for 'empowering participants to take ownership in guiding the growth of the project' (Parilla and Ferriter 2016, p. 441). Transcription is recognized as an act of subjectivity, where diverse interests and varied foci well may color the interpretation of the records to be transcribed (Pichler 1993). However, despite positive attitudes toward users and a widespread appreciation of the need for communication, the focus of the project owner is often on giving authoritative, unidirectional instructions rather than on fostering mutual dialogue. Even in projects that support discussion for a connected to the transcription work-like the Old Weather project (Blaser 2014)-little is mentioned about volunteer influence over the structures or conditions of transcription.
To gain a better understanding of how volunteer influence is exemplified and discussed, one must study the bigger picture of participation in archives. A corpora of texts refer to the role of 'others as archivists'-meaning that non-professionals are allowed to participate in curatorial work with archival records and to cooperate with archivists in managing, describing, and organizing them (Huvila 2015, p. 369). Such participatory curation activities are pointed out as especially essential for archiving records documenting governmental decisions about individuals and used to promote governance frameworks for participatory decision making and recordkeeping (Evans et al. 2019). Here, the Participatory Records Continuum Model serves as a refined example of how the level of influence correlates with the distance of the participatory agents' relation to the activities that the records represent. This relationship adds a theoretical situatedness to participatory agency where a record with a more central role for a participant advocates more agency compared to records that are relevant in a more peripheral context, like for example research or education (Rolan 2017). While this model presents a carefully packaged argument for implementing archival involvement, other approaches to interaction between archivists and 'the others' are often less holistic when it comes to balancing the interests of participants against the need to secure archival custody.
Some early examples of co-design draw attention to the essentiality of relationships based on an 'equal footing' between volunteers and archivists, to overcome 1 3 power biases and minimize the risk of divisiveness within projects (Srinivasan et al. 2009, p. 272). Furthermore, it is argued that listening to the community and caring for its interests in participatory processes is important for making sustainable participation possible (Baxter 2011) and for strengthening other relations with archival audiences (Carletti et al. 2013). These approaches all speak in favor of appreciating volunteers and developing constructive relationships with them; however, they do not say much about how this should be achieved.
Sociability and two-way communication between project initiator and community have proven to generate participatory projects that, compared to projects that could not be characterized as sociable, have been more sustainable and successful (Liew 2015). Communicating with participants, it is explained, serves to increase their motivation and trust, whereas a lack of communication can result in unsuccessful and unfinished projects (Parilla and Ferriter 2016). Given the stated importance of the communicative aspects of participation, different models have been suggested for contextualizing processes of participatory influence and negotiation, each with its own take on the relationships between participants and professionals.
From the pedagogical field, the Reggio Emilia model puts emphasis on dialogue between an authoritative part and a less authoritative one; for example between professionals and amateur participants. The model criticizes the traditional hierarchical role between teacher and student, arguing that the needs and inner motivation of the participant should be the driving force behind any project and that volunteers should be treated 'as equal authorities' (Phillips 2014, p. 261).
Yet another model of archival participation is the 'Rhizome'. This vision emphasizes the need for non-hierarchical and a centric interactions between the parties involved (Duff and Haskell 2015). A rhizome, namely, has no clear beginning or end, it is both flexible and large, and it co-exists with other structures without dominating them. However, while this model suggests how participatory influence is to be incorporated into archival structures, it does not address how input from participants is to be documented and organized.
In relation to the Reggio Emilia, Rhizome, or similar models, community-based participatory research (CBPR) offers a structured framework for participatory work that highlights the mutual advantages of participation. It ensures that collaboration benefits both parties without compromising the aims of the project or its prospects for success. CBPR thereby increases the chances for an ethically balanced and successful project. Central features of CBPR include mutual learning, mutual decisionmaking, mutual ownership or use of resources, the enhancement of participants' capacities, and volunteer influence at the early stages of a project (Coughlin et al. 2017). Popple and Mutibwa have shown how CBPR encourages engagement and optimism on the part of participants when an online platform is developed for collecting stories and memories and engaging with cultural heritage material (Popple and Mutibwa 2016). The two authors also highlight the importance of a dialogic space for discussion, devoted to negotiations and reflexive processes. Such a space (either physical or virtual) provides a contact area for all of the partners involved in a project. Its essential role in collaboration has been illustrated in studies like that of Smith and Iversen (2014). It can consist of group meetings, advisory boards, or online discussion forums.
CBPR is a tool that facilitates working with the most problematic part of archival participation: balancing participatory influence with archival authority. However, if we are better to understand the power dimensions of participation and the need to allow volunteers to influence the conditions for their participation, we must grasp the theoretical principles of participation.

Observing the negotiation of influence with a CBPR approach
Participation comprises multiple forms and levels of involvement (Pateman 1970). By scrutinizing the notion of 'participation', we can see how the tension between on the one hand participant involvement in situations where the framework is already set, and on the other hand the involvement of participants in situations where they have a structural impact on the conditions and circumstances of their participation, is an issue not isolated to the domain of cultural heritage. It is also important in other contexts where democratic participation is sought. However, the depth and authenticity of participation is always dependent on the opportunities that participants have to exert influence and to make an impact. If the concept of 'participation' is to be a useful tool for analyzing influence, it needs to be refined, so as to make it possible to distinguish forms of participation with diverging potentials for participant agency.
Maximalist and minimalist participation provide a means for identifying the space for empowering participants, while at the same time achieving the other aims of archival work. Maximalist participation represents full and authentic participation, where democratic structures enable participants to influence the terms of their engagement. By giving participants the opportunity to control the situation (at least in part), maximalist participation empowers volunteers and allows them to engage in the work on their own terms (Carpentier 2015). Minimalist participation, on the other hand, implies a weaker form of influence: It may entail representation only, where volunteers can only take part in a shallow way, without being able to influence the structures of the participatory situation (Carpentier 2015).
In an archival context, the sharpest conflicts with the traditional authority of archivists supposedly arise in situations marked by maximalist participation. However, maximalist participation presents archivists as benevolent gatekeepers (Gauld 2017) who grant participants the ability to modulate their engagement according to their aims and needs. Maximalist participation is therefore a useful concept for framing essential elements in the ambiguous situations of conflicted influence and power that this article seeks to investigate.
To identify situations of maximalist participation and operationalize their dynamics, this study uses a CBPR-inspired framework as an analytical lens. The elements of this framework will now be presented in more detail.

A CBPR-inspired framework
Due to its ability to include the voices of marginalized groups, the CBPR approach has been popular in the health sciences for more than 20 years (Coughlin et al. 2017;Wallerstein and Duran 2008). The definition of CBPR principles used in this analysis is based on an overview by Coughlin, Smith, and Fernandez (2017). These principles are presented below, clustered by their association with shared influence, mutual development, and mutual use.
1. Shared influence signifies the first cluster. It contains two different CBPR elements: participatory influence in the early stages of a project, and mutual decision-making. This part of the framework emphasizes that volunteers should be involved already in the planning phase of a project, and that their needs and priorities should be considered and assessed. This approach should then permeate the project continuously from that point on. Evaluation of the priorities and agendas of the community and of researchers should involve representation for both parties. The 'dialogic space' is an essential tool for realizing shared influence. 2. Mutual development comprises mutual learning and the enhancement of the capacities of participants. Co-learning, mixed perspectives and experiences, and the democratic creation of knowledge are given space. The notion that participants and experts have different knowledge and capabilities is central here, as is the conviction that all experiences are valuable for the development of a project. These priorities can take the project in unexpected directions-which is why it is important to expect the unexpected, so that it will be easier to acknowledge such moments of mutual learning as constructive, instead of regarding them as instances of negative conflict. Building up the capacities of participants means allowing them to develop skills and abilities along the way, thereby enhancing their self-sufficiency when they undertake similar activities outside the project. 3. Mutual use incorporates mutual ownership and mutual use of resources into the analytical framework. These components of CBPR are meant to ensure that the final result of the project is made relevant to and beneficial for the community (Coughlin et al. 2017). Depending on the type of project, such elements may be manifest in different ways-e.g., in the shared ownership of research products. In cases where the results consist of new knowledge or other intangible values, the focus is more on granting access to the results which the project yields. In other words, results must be disseminated and made accessible for community use. Moreover, the results should benefit participants and be useful to them-which will likely increase their motivation to engage in the project.
Taken together, the different elements of the framework support the development of trust among the agents active in the project. If the criteria set out in the CBPR-based framework fail to be realized-e.g., if the results are inaccessible to the community or irrelevant from the standpoint of its concerns-it might instead cause archivists to lose the confidence of volunteers (Hatch et al. 1993).

Material and methodology
The online user forum that supports Begravelser 1861-1912 acts as the main arena for communication and negotiation between participants and archivists in the transcription project. It consists of seven main sections.
Opinions about database content and design that are of interest for this article are mostly found in the section 'Feedback on Kildetasteren', which is why this part of the forum was selected for analysis. The number of posts selected for analysis covers all content from the start of the forum in November 2016 until September 2017. The forum is still active (as of June 2019), and new posts are submitted continually. At the end of September 2017, the feedback section consisted of around 1300 posts on some 300 topics.
Of these 1300, only posts that discussed user influence as defined by the CBPR approach were selected for analysis. Prime emphasis was placed on arguments for transformative change to the database content or functionality, such as re-organization or adjusted definitions of data categories. Every post that was deemed relevant was coded to reflect the type of database modification that had been suggested. Posts addressing the same topic were then grouped together, so as to make it possible to follow the negotiations concerning specific suggestions through the whole cycle-from the time a suggestion was first made to the point where it was announced that said suggestion either had been rejected or would be executed. Employing the framework, it was possible to identify several discursive themes of change and negotiation that could be connected to CBPR.
To understand the progression and development of the project, an e-mail interview was conducted with one of the archival administrators of the forum, and project documentation and open project newsletters posted in the user forum were analyzed. The newsletters contained announcements by the City Archive of changes in the funeral database. Updated newsletters were published in the forum on a monthly basis, making it possible for an observer to follow how the database expanded as time passed.

Limitations
Suggestions for improvement could also be submitted via email, or other communication channels but these were not included in the analysis. However, according to the archivist interviewed, the contents of the forum can still be considered representative of the negotiation process regarding the database and its structure (Interview, personal communication, 16 November 2017).
In the first months of the project, only a small number of participants introduced new topics in the forum or made suggestions about the database. The number of participants active in the forum during the period of the study was lower than 20 throughout, although which specific individuals were active changed over time. While the small number of participants can be seen as a limitation of the study, it accords with the findings of other studies of online participation-namely that most of the work is done by a minority of volunteers (Dunn and Hedges 2012).

From transcribers to database designers: How participants gain influence
This section contains a review of the points of maximalist participation, as well as a discussion of how participants influenced archival practice and transformed the ultimate transcription in the project, beyond the original design.

Shared influence
In regard to its functionality in the project, the forum answered well to the dialogic space emphasized in CBPR. The argumentation and negotiation around some of the modifications transformed the database into a more useful product for the transcriber community. This indicates that a constructive and mutual learning process took place, where both participants and archivists had a hand in shaping information structures and conditions of work. However, the forum was not activated until the transcriptions had already started, meaning that the perspectives of participants were not represented through this channel. Most participants did not seem to react to the loss of this opportunity; however, one volunteer did criticize the initial simplification of age-of-death registration (approximated to the month instead of the day), saying that: 'One wishes one had been given the opportunity to test the transcription tool before [the registration started]; one simply wishes one had been given the opportunity to participate (#1431).' 3 On the other hand, volunteers from an earlier transcription project at the City Archive had been consulted during the initial phase of the project and had been able to influence its design (Interview, personal communication, 16 November 2017). However, this early engagement of non-professionals was not formalized in any reference group, and it is not documented how these persons influenced the project (Copenhagen City Archive 2017a, b; Interview, personal communication, 16 November 2017). It is also hard to say how well they represented the perspectives of the subsequent participatory community, since that community was formed after the project began. A participatory community only emerges through the act of participation. This makes it difficult to engage the community before a participatory project begins, since the participants cannot be accurately identified. It can thus be hard to meet the CBPR requirement that participants be included from the absolute start of a project.

Mutual development
The discussion of problems in the forum encouraged participants to learn more about the historical records they were transcribing, and it prompted them to think more thoroughly about problems and possible solutions-one of the advantages of forum discussions (Holtz et al. 2012, p. 56). Moreover, by responding to participants' questions, the archivists learned more about the records as well, in some cases motivating them to make adjustments to the database. The forum thus enhanced understanding of the records, as well as appreciation among both archivists and participants of the complications associated with structural transcription.
As participants enjoyed support from the forum community, they felt encouraged to find solutions to very practical problems, such as how to make out handwriting. They also felt emboldened to seek a deeper understanding of the historical context and conditions of life surrounding funeral practices, as in discussions about unusual causes of death. Their skills thus enhanced, many participants were also better equipped to pursue their own genealogical research, increasing their motivation to take part in the project.
Participants that showed outstanding skills and commitment to the project could also be 'upgraded' to super-users, granting them a semi-professional status in the forum. These participants were given administrative responsibility for one of the vocabulary lists (causes of death, graveyards, occupations, etc.), and they were also expected to assist the archivists in answering questions (Interview, personal communication, 16 November 2017). Thus, appointing super-users from participants who had been 'especially active, dedicated, and helpful in the forum' served to bridge the gap between participants and professionals, and to delegate some of the work and responsibility to participants (Interview, personal communication, 16 November 2017). Such social rewards are believed to increase intrinsic motivation, enhancing the capacities of such persons and cementing their loyalty to the project (Dunn and Hedges 2012, p. 10).
The case of the super-users showed that increasing one's skills and capacities as a participant paid off, in that the archivists showed greater confidence in volunteers who had done this. Super-users thereby achieved a status closer to that of the archivists.
As mentioned earlier, the forum helped both participants and archivists to gradually gain a deeper understanding of the records. Two cases will now be examined, to show that enhanced understanding does not automatically mean consensus over priorities. The aim of the project was to register all information concerning the deceased person in the register post, but to keep information about family members or other relatives to a minimum. Some volunteers argued, however, that the relations of the dead person were a matter of great interest. As one participant put it: 'There is a lot of information to be gotten from the words 'father', 'mother', 'spouse'' (#98). 4 Another pointed out that registering the names of the parents of dead children was essential if connections within families were to be found: It is very important to register relations of the dead person. That identifies the dead. For example, genealogists do not know the name of a dead child in most cases, but the child can nonetheless be found with the name of the parents (Participant, #101).
One example of this concerned cases of illegitimate children. A particularly vivid debate ensued. During the initial stages of the project, the full name of the mother of a dead illegitimate child was not registered in the database, despite its having been stated in the funeral records. The first name of the mother could be noted in the free-text commentary field (although it could not be retrieved through a search in the name field). According to two participants, this meant that research opportunities were being disregarded-in connection, for example, with the effect of twin births on the later number of children in a family (#457, 686, 699, 702, 717). As one volunteer noted: On 80 pages, I found 10 unregistered mothers of illegitimate children. They even had different last names.
[…] How can you find the twins? (Participant,#730) Despite this remark, the archivists argued that such cases were too few to make any difference. They shut the discussion down. The person who had made the suggestion felt irritated at this, and did not agree that the method of registration made for sufficient search opportunities. The participant in question, who had been highly engaged from the start, later stopped being active. This discussion (among others) resulted in misunderstanding and diminished motivation.
In one case, a shared deeper understanding of the records actually resulted in transformation of the database. It had to do with the registration of the relationship between persons and occupations. While the professionals were content with just making a lexical list that declared the occupations mentioned in the records, many transcribers were interested in the relationship between a person and an occupation. The background to this problem was that an occupation was often recorded as an identifier of the dead person, regardless of whether a person was actually involved in that occupation. Thus children, spouses, and foster children were often registered as having had the same occupation. Furthermore, even in cases where the occupation had been practiced by the dead person, the relationship of that person to the occupation may have changed over time, due to retirement. After having highlighted various cases of this kind, transcribers were able to introduce13 different relationships between persons and occupations into the database (spanning from 'own occupation' to more distant relationships like 'former occupation of father') (#4065).

Mutual use as a driver of database modifications
The CBPR framework also shows how the forum made the database a more useful and relevant information resource for participants, professionals, and others(e.g., academic researchers). Thanks to the discussions in the forum, the archivists expanded the database so that it covered more information categories. They also increased the exactness of existing categories and defined the relationships between categories better (see Jansson 2018 for an extensive description).
Most of the arguments in the posts concerned data that could be found in the records but which, for various reasons, had not been included in the database from the beginning. Some participants saw the archivists' choice of data categories as less than optimal, and they used the forum to express their doubts: I don't understand your choice of data. I think almost everything should be registered. There is not that much data in each funeral record, and it is noted in a very simple way. It is fine not to register the prices, and the funeral dates may not be that important, but the rest is relevant. (Participant, #254) One reason why the archivists limited the data to be transcribed at the beginning of the project was to avoid scaring participants away from the transcription project by presenting them with an extensive registration form. This caution later appeared misplaced, inasmuch as the exclusion of information upset participants, because they felt they could not do the job as meticulously as they wished to in order to create an information resource useful for their needs.
Some examples of categories that were introduced or adjusted as a result of suggestions from participants were Serial number, Institution/Place, Place of death, and Comments.

Introducing new information categories
As mentioned, one of the new information categories was serial number. When a participant first suggested it, an archivist replied that it was unnecessary and would not be added (#55). However, as the project went on, a series of participants made the same suggestion (#44, 57, 59, 90, 91, 97, 177). They argued that they needed a navigator with which to orient themselves among the register posts. Serial numbers would allow them to go back to posts if necessary and to connect database instances with records. After some debate, the new category was finally included in the database (#144). This is one of the first examples of where participants managed to influence the contents of the funeral database so as to grant achieve better working conditions for themselves, as well as to make it easier for archivists and researchers to retrieve the original record behind a given registration.
Another example concerns the categories of institution/place and place of death. Information about institution or place (meaning the place where the deceased person had lived) was used to describe the many cases where the records did not specify a street address (as for hospitals, houses of correction, factories, or prisons). The category of institution/place, like that of serial number, was suggested quite early on and by various participants (#171, 254). The suggestion was also received much more positively by the archivists, who introduced the new category without much fuss [Nyhedsbrev 4, #603 in Nyhedsbreveog information frastadsarkivet (Newsletters and information from the City Archive)].
Cases where two funerals were fused in the same register post (as when twins passed away just a few days apart) were pointed out as a complication by a handful of participants (#619, 695, 697). As a result, just one person would be listed in the database when actually there were two. Apparently, this problem had not been considered in the database design. Although recognizing it as a problem, the professionals argued it was of little importance and that it would require disproportionate effort to fix, since they believed it only concerned a few cases: The structure of the database is designed in accordance with a one-to-one-relationship between funeral and person. It is a big change to make now, and we have many wishes on the list, for example [regarding] the search function and the statistics. But we are aware of the problem here, and we realize it hides other persons that can be connected with a funeral. It is our impression [, however,] that this only concerns a few persons […] (Archivist, #626).
On the contrary, one participant, in particular, argued the inability to separate such fused posts was a serious problem. The way in which the archivists approached the problem seemed to undermine this participant's engagement. In a response posted in the forum, this person pointed out the gap in interest between professionals and participants, and seemed to express distrust in the former: Few participants made such an outright accusation, but the words above do illustrate a sharp difference in focus between the two groups. The participants often desired more specific information. One reason for this lay in their genealogical interest, which made them more concerned about individuals than about statistical correlations. Another lay in their serious commitment to the project, and their feeling of responsibility toward it. Yet another reflected their expectations in connection with transcription. The widespread view among participants was that 'transcription' referred to a literally accurate reproduction of the information in question. This prompted the archivists to write a note explaining that the aim of Begravelser 1861-1912 was to produce structured transcription with normalized data (Nyhedsbrev #12, #2492).

Debating free-text comments
Already at the beginning of the project, participants called attention to the lack of any opportunity to add annotated observations or clarifications about a funeral (#75, 158). This could involve references to contextual data, external information sources such as church registers, or annotations in the margin of the record about such things as where the deceased was found. Other interesting data that participants mentioned as just being passed by without any space being allocated for them included names, addresses, suicides, foster children, and close relationships with other persons. However, the City Archive rejected suggestions for a dedicated field for user comments, on the grounds that this would risk introducing unstructured information into the database, making it too messy as an information source: We are quite reluctant to [include] a commentary field, as experience shows that it turns out to be difficult to control the contents of such an information field (Archivist, #207).
Information of this kind is also unsuitable for quantitative or automated analysis. Such automated data analysis is mostly done by academic researchers working in large projects with substantial technical resources for collecting and analyzing data-resources to which amateur researchers rarely have access. Academics thus have little interest in the type of information in question. Over time, however, some persuasive transcribers managed to soften the professionals' reluctance, using such arguments as: 'If there are good guidelines for the commentary fields, we will surely be able to use them sensibly' (Participant, #213). The archivists then promised to consider comment fields: We will evaluate the need for a commentary field. You are welcome to send us an email with examples where you think comments would be relevant. Then we will have a look at it. (Archivist, #260) The following response from the participant included a list of the advantages that comment fields would provide for genealogists. It ended by stating: […] This is important.
[…] As a genealogist, one does not know the names or birthdays of dead children, but one can see from the census of 1901 if there are dead children. Nor does one know the addresses of all one's ancestors. All information can be useful in genealogical research (Participant, #264).
With arguments like this, participants finally managed to persuade the archivists to include a comments field. The struggle may have reflected the differing expectations of archivists and participants. It might have been possible, moreover, to resolve it through a better implementation of the mutual decision-making for which CBPR calls. The initial design of the database might then have been more closely attuned to the needs and interests of genealogists. Professionals had first expected the discussion forum to be a place for discussion and miscellaneous annotations; however, users were not content with this ephemeral solution, which lacked sufficient connection to the database. They identified a need for more sustainable annotations that would be saved in connection with register posts. Including this kind of miscellaneous information in the database thus recognized the learning process of the transcribers, as well as honoring their work and effort in checking names and other facts in external reference documents (like church registers and encyclopedias).
In conclusion, the comments did more than just enable participants to make the information in the final version of the database more useful from their standpoint. They also increased their motivation to take part in the transcription process.

Conclusions
Although Begravelser 1861-1912 was not expressly designed as a CBPR project, this study shows that it more or less followed the main principles of CBPR. It both allowed and encouraged participants to influence the conditions for their participation. Compared to many other studies on participatory-transcription projects, this article highlights the advantages of maximalist participation. Furthermore, it does more than just point out a method for more ethical participation; it also demonstrates that a benevolent approach to maximalist participation enhances the relevance and quality of such projects. Two main conclusions can be drawn from this study: First, a maximalist participatory process need not conflict with the need for archival custody or with the aims of a transcription project. An online discussion space like the forum analyzed in this article invites participants to express their needs and desires in a regulated way. Archivists can retain executive power over the project, even while assuming the role of benevolent gatekeeper. CBPR(as well as other approaches that stress the need for collaboration) enables archivists to act in an ethically more responsible manner, by allowing alternative perspectives to be represented in archival collections-while at the same time standing guard over institutional knowledge and crucial information. However, the difficulties associated with maximalist participation in an archival context need to be recognized as well, as the results of this study indicate. For example, the debate over the comments field highlighted considerable institutional rigidity vis-à-vis the changes that participants proposed in project plans. The suggestion for such a field was only accepted after thoughtful consideration and energetic efforts at persuasion by several participants. Thus, even when participants are allowed to influenced the conceptual content of a database and its functionalities, the archivists can still refrain from implementing the changes proposed. In relation to the Participatory Records Continuum Model (Rolan 2017) discussed above, the transcribers have a weak relation to the records that are being transcribed, but what is worth underlining is their strong relation to the database, as a product of their own work. It exemplifies how participation itself builds new relations to the records, which is why agency cannot solely be dependent on participants' relation with the archival record (as the model suggests) but also with their engagement in the work which participation enables.
Applying a CBPR approach also necessitates careful planning of time and resources, so that the ideas and proposals of participants can be properly considered and given their due. The risks arising from failure in one of the areas of CBPR also need to be borne in mind, as such failure may result in diminished trust for the archival institution.
Second, as the forum discussions reviewed above make clear, mutual decisionmaking encourages a deeper understanding of records and databases. Mutual learning enhances the insights of both participants and archivists; knowledge acquired through participation produces synergetic effects. Furthermore, the development of skills generates more qualitative results as well. In the Danish case considered here, for example, the database was adjusted as an information resource so that it better reflected the content of the records. The potential usefulness of the database was also enhanced, inasmuch as a better understanding of the records helped reveal research possibilities and information needs of which the archivists were not aware when the project began. It may further be argued, in view of the experiences reviewed in this study, that maximalist participationespecially on the basis of a CBPR approach, with its explicit requirement of mutual development-is important not just for ensuring participatory influence, but also for promoting innovation.
The CBPR approach does not solve all problems of participation and representation in connection with archives. Even if this approach is meticulously followed in a given project or process, the context and organizational structure surrounding a project can still limit or entirely nullify the effects of maximalist participation (especially when it comes to archival representation). However, CBPR does offer a functional approach to the 'power sharing' (in Dahlgren's words) which a truly ethical and responsible archival practice requires if its participatory procedures are not to deserve the label of 'fraudulent'.
For CBPR to be applied successfully in archival work, we need a greater understanding of participatory influence. More research is needed on how the values and perspectives of diverse user groups are prioritized by archivists, and on how these perspectives influence the dynamics between archives and their users in processes of shared curation. Finally, there is a need for research that keeps questioning the relationships between institutions and their audiences-especially since there are no signs 'the participatory turn' will end anytime soon.