Abstract
Within the digital humanities, there are many approaches to citation. Every
discipline handles citation and authorship differently, and within the digital
humanities there are wide divergences of practice in the ways that credit is
assigned and made visible. However, there is also broad agreement that citation
practices need scrutiny and perhaps rethinking. The issue arises with particular
force when we consider how to cite digital humanities projects and tools.
Standard citation practices do not provide good precedents for making visible
the contributions of project personnel in these highly collaborative efforts.
There has been significant informal discussion and debate in recent years
concerning appropriate ways to credit this work, but no consensus has been
reached. FairCite (http://faircite.wordpress.com) was founded to promote this discussion
and encourage it towards practical, public outcomes.
Citation is not just about information management, ensuring a reader can find a
particular item at a future date. If that were the case our footnotes and references
would consist solely of the numerical International Standard Book Numbers (ISBNs)
attached to every book, or the Digital Object Identifiers (DOIs) and Uniform
Resource Locators (URLs) of the online world. Within the ecology of scholarship,
citations also provide a verifiable way to track the contributions of our colleagues
and competitors through authorship in the broadest sense. In this role, citation
provides a record of achievement and, in the emerging world of linked data, a view
of the interconnected intellectual landscape of scholarship. However, the
traditional list of authors is not an exclusive inventory of all of those people who
had a hand in the creation of research outputs. No academic field currently takes a
radically inclusive approach to authorship that would include everyone who had any
involvement whatsoever in a project. And arguably, such an approach could prove
theoretically as well as practically untenable. Once we begin exploring the
relations of dependency thoroughly, we encounter a potentially limitless set of
roles that may have influenced the final form of a work, such as peer reviewers,
seminar colleagues, conference respondents, copyeditors, and many others. And once
we extend our understanding of “authorship” in the digital humanities to include
project and tool development, the set of relevant participants in the creative
ecology grows even further. In designing citation systems that do justice to the
record of achievement, we must ask how the threshold of citable intellectual effort
and responsibility is constructed and justified.
For the digital humanities, we do not imagine that there can be a single answer to
this question. Every discipline handles citation and authorship differently, and
within the digital humanities there are wide divergences of practice in the ways
that credit is assigned and made visible. However, there is also broad agreement
that citation practices need scrutiny and perhaps rethinking. The issue arises with
particular force when we consider how to cite digital humanities projects and tools,
since for many alt-ac professionals, projects are a crucial professional currency
just as publications are for academic faculty. However, standard citation practices
do not provide good precedents for making visible the contributions of project
personnel in these highly collaborative efforts. There has been significant informal
discussion and debate in recent years concerning appropriate ways to credit this
work, but no consensus has been reached.
FairCite (
http://faircite.wordpress.com) was founded to promote this discussion and
encourage it towards practical, public outcomes. Initiated by Adam Crymble in 2011,
FairCite acknowledges that scholarship is increasingly interdisciplinary and team
based, and traditional models of authorship need to be redefined in light of our
changing work environment. Without this redefinition, the concept of authorship
itself will gradually cease to be a meaningful representation of contribution to a
project. FairCite also provides a reminder that the career development of alt-ac
scholars may depend on the strength of their C.V.s and portfolios. These members of
our project teams can ill afford to have their contributions concealed by
traditional academic practices that exclusively highlight the role of the senior
scholar, or by a “no names” policy in which no one is explicitly treated as an
author.
[1]
Because academic disciplines tend to be insular, it is easy for a scholar accustomed
to solo-authored contributions to get the impression that shared authorship is the
exception rather than the norm. However, other models of credit are used throughout
academia to acknowledge the various contributions of collaborators with few or no
detrimental effects. This article highlights some of the models currently in use,
and provides practical guidance for both principal investigators and alt-ac team
members looking to engage in a productive dialogue with colleagues to ensure that
the needs of all parties are considered and a fairer culture of credit can begin to
emerge.
FairCite does not intend to define at what level that contribution becomes “enough”
to warrant authorship. FairCite neither presumes nor suggests that all members
involved in a collaborative project should be considered full authors of the
outputs. Neither is it about padding the C.V.s of those most in need of career
advancement. Rather, it is about being informed about the needs of all members of a
project team and engaging in an open and honest conversation with them to ensure
that credit is shared in a way that represents the contributions of those involved.
This paper was written with the example of a team of individuals involved in
creating a web-based digital humanities project in mind, but it is hoped that the
messages will be general enough to apply to a range of academic situations,
including those resulting in a jointly-authored journal article or a monograph.
Existing Models
The humanities have few models for collaborative work. The single-authored paper
or monograph is still expected to be the humanities scholar’s predominant
output. However, there are some precedents and useful guidelines in a number of
related fields including digital humanities and academic sciences. The Wikipedia
article on “Academic Authorship” also offers an excellent overview of different
disciplinary philosophies towards sharing credit [
Wikipedia, Authorship].
Digital Humanities
In the majority of cases it is left up to the reader to determine how to cite
an academic website or digital project. Sometimes websites will provide
advice for how best to cite the work; the
Old Bailey
Online is a good such example, which offers suggested citations
that can be used in a range of contexts depending on whether or not one
seeks to cite the project itself, a document stored in the database, or even
a search result. While this is incredibly useful it is by no means an
industry standard to include such advice. If this information is available,
it is typically found on the “About” or “Project Team” pages. At best it
will provide a suggested citation, and will outline the names and
contributions of all members of the team. This style of credit is not unlike
what one would expect at the end of a film in which the cast and crew are
listed along with their job titles or characters.
In the autumn of 2011, when the first discussions about FairCite began, they
were centred on the
Old Bailey Online. At the
time the project’s policy was to include no one’s name in the project
citation:
Old Bailey Proceedings Online. Version 6.0,
March 2011. http://www.oldbaileyonline.org/
The principal
investigators had come to this policy as a way to avoid concentrating the
credit for such a large project (one that included over 40 collaborators)
into the hands of a few. This was done with good intentions, but meant the
names of technical team members and project managers were rarely if ever
given prominence. After a series of discussions between Tim Hitchcock, one
of the project’s principal investigators, and Adam Crymble
[2], the
Old Bailey
Online team decided to change its suggested citation to:
Tim Hitchcock, Robert
Shoemaker, Clive Emsley, Sharon Howard and Jamie McLaughlin, et al.,
The Old Bailey Proceedings Online,
1674-1913 (www.oldbaileyonline.org, version 7.0, 24
March 2012).
[Emsley et al. 2013]
The resultant citation was determined by the members of the project
team, rather than being imposed from outside. There was no precedent for
this type of output with a suggested citation that takes authorship into
account, so there was no right or wrong answer. What is most important is
that the team engaged with the needs of its team members and came up with a
customised solution that they believed fairly represented the intellectual
contributions of everyone involved. In this case the five names represent
the three academic principal investigators, the project manager, and the
lead technical developer. Thirty-five other team members are represented by
the “et al” at the end of the list of names, and are acknowledged more
formally on the project credits page.[
Emsley et al. 2013] The
solution that the
Old Bailey Online team came
to meshes well with the
Collaborator’s Bill of Rights
[
Kirschenbaum et al. 2011], which states that “All collaborators should feel empowered to
express their contributions honestly and comprehensively.” The
Bill was the product of the
Off the Tracks
workshop held at the University of Maryland, an event that sought to promote
the professionalization of the digital humanities.
Thus far, models of credit in the digital humanities have tended to follow
the disciplinary standards of the department into which people are hired.
There are as yet few digital humanities departments; most colleagues work in
either digital humanities “centers”, cultural heritage organizations, or
traditional academic departments such as history or literature. However,
when disciplinary standards in the hiring department differ widely from
those used by digital humanists, tensions can mount. Zotero, the citation
management software developed at the Roy Rosenzweig Center for History &
New Media at George Mason University, lists contributors on an About page.
The project does not suggest anything resembling a citation (despite,
perhaps ironically, being citation management software).[
Zotero] For the project’s director, Sean Takats, the
intellectual distance between his contributions to Zotero and the monographs
his colleagues were used to evaluating led to conflict during his tenure and
promotion evaluation, which Takats shared on his blog. One of the members of
the committee questioned whether involvement with Zotero “constitutes actual research (as
opposed to project management)”
[
Takats 2013]. Sean’s ire at having such a successful project
marginalized by colleagues ill-equipped to pass judgment on something so
different from the work they produce themselves is certainly valid. However,
it is perhaps because projects like Zotero are so different that it is
important for digital humanists to come up with models for sharing credit
that elucidate the nature of everyone’s contributions in a way that is easy
for outsiders to understand.
As Bethany Nowviskie’s noted in “Where credit is
due”, a talk delivered at the NINES Summer Institute in May 2011,
the problem of being recognized does not stop at those seeking tenure, but
permeates into the much more vulnerable realm of alt-ac scholars in the
digital humanities. Nowviskie worries that these scholars are put “into a position where they
may choose to de-emphasize their own innovative but collaborative
work because they fear it will not fit the preconceived notion of
valid or significant scholarly contribution by a sole
academic.”
[
Nowviskie 2011] The way out of this predicament may not be led by individual scholars
interested in promoting their own career progression, but in a change in the
system that rewards sharing of credit.
In the United Kingdom the research of scholars, departments, and whole
universities is ranked and subsequent government funding to the institution
is based on the results. The ranking occurs once every six years. In 2014
the exercise is known as the
Research Excellence
Framework (REF), and this version of the assessment emphasizes the
importance of digital outputs such as databases, digital media, and
software, which will be considered a form of scholarly output and will be
assessed alongside journal articles and monographs. This means that in the
UK at least, scholars will have a vested interest in including their names
with project outputs and the “no-names” policy may disappear out of a need
for career advancement and departmental ranking.
Academic Science
The sciences are known for multi-authored papers and provide the most
comprehensive model for author attribution. However, there is no single
model and each journal may have their own guidelines. Nevertheless there are
several attempts at standardization. The
International
Committee of Medical Journal Editors (ICMJE) supports the
following standards and expects authors to meet all three conditions:
- substantial contributions to conception and design,
acquisition of data, or analysis and interpretation of
data;
- drafting the article or revising it critically for important
intellectual content; and
- final approval of the version to be published.
[ICMJE 2013]
Journals under the Royal Society of Chemistry Publishing umbrella provided
the following guidelines:
There is no universally agreed
definition of authorship. As a minimum, authors should take
responsibility for a particular section of the study. The award of
authorship should balance intellectual contributions to the conception,
design, analysis and writing of the study against the collection of data
and other routine work. If there is no task that can reasonably be
attributed to a particular individual, then that individual should not
be credited with authorship. All authors must take public responsibility
for the content of their paper. The multidisciplinary nature of much
research can make this difficult, but this may be resolved by the
disclosure of individual contributions.
[RSCP 2013]
The Nature family of journals does not
explicitly outline criteria for authorship, but requires all papers to
detail the contribution each author made in a footnote. This policy is now
mandatory for all Nature papers and is meant to
eliminate the problem of “Honourary Authors” – people
whose names are included on the paper despite no tangible input from that
person. Nature also believes this policy
ensures that the right person can be made responsible should problems with
the research arise in the future. Examples of author contribution statements
from Nature papers look like the following:
- S.H.C. designed and performed experiments, analysed data and wrote
the paper; N.C., M.T. and J.M.G. designed and performed experiments;
D.R. and M.B.G. developed analytical tools; and C.I.B. designed
experiments, analysed data and wrote the paper.
- Y.O. and Y.Z. designed the experiments and prepared the
manuscript. Y.O. performed the experiments. G.S., M.K.R. and Y.M.
generated the chimaera mice from the BayGenomics ES clone.
[3]
Scientific papers commonly include a footnote attached to the author list,
which provides details on where each person works and how they can be
contacted, as in the example below from
Nature.
An equivalent to this footnote space could be used in digital humanities
projects to outline contributions from each person. The wording could be
left up to individual projects, or could be standardized with a limited set
of contribution keywords (writer, designer, database, project manager, etc).
Nature Editors have contributed a number of
editorials discussing the issue of authorship, many of which are relevant to
the current discussion; see [
ICMJE 2013], [
Clarke 2007], [
Nature Nanotechnology 2009], [
Nature Cell Biology 2009a], [
Nature Cell Biology 2009b], [
Nature Physics 2009], [
Nature 2007], [
Nature 2009a], [
Nature Materials 2004], [
Nature Materials 2008].
As not all people who participate in a project meet the criteria for
authorship, it is common in science papers to
“acknowledge” more minor contributors in the same way
one acknowledges funding bodies. The ICMJE provide guidelines for
acknowledgement, which are similar to their guidelines for authorship, and
outlines the appropriate model for acknowledging said contributions. For
example:
The authors would like to thank
Ms. Anne-Marie Allen and Mr. Adam Crymble for their assistance with data
collection and the Natural Sciences and Engineering Research Council for
their financial support.
[Kedgley et al. 2009]
Having a Discussion about Credit with Collaborators
FairCite is not an attempt to enforce a standardized system of liberal author
attribution and citation. Instead, it is about promoting an open and honest
conversation between collaborators. This section outlines one way in which
project team members might like to initiate and conduct that discussion. This is
by no means the only way to do so but may prove useful for getting the ball
rolling.
Strategies for Initiating the Discussion
The biggest hurdle to overcome is to start the conversation in the first
place. The sooner this conversation takes place the better. Ideally it
should be part of the original project pitch in which the team is being put
together and the outputs are discussed. By being open and upfront about the
intended project outputs and how credit is shared there is less opportunity
for resentment or awkwardness to build up that could threaten the
productivity of the group as well as future relations between its members.
The person in a position of power is the best placed to instigate this
conversation, since more junior members may be hesitant to raise what they
may think is a contentious issue. Usually this is the principal investigator
or the individual who first suggested the collaboration.
However, it may not occur to that person that a conversation is needed, and
so all collaborators should feel empowered to raise the issue and seek a
discussion. Many people will be relieved to have the question raised and
dealt with, even if they themselves were unwilling to raise it. At worst the
team discovers at an early stage that they do not want to continue working
together and everyone has the chance to cut their losses. At best you avoid
an awkward situation and hurt feelings down the road, and everyone’s career
benefits from the outputs of the project. If your project is already
underway and your team has not yet had this conversation, it is not too
late; but the sooner it occurs, the better. If your group has regular
meetings consider raising it at the next opportunity. Alternatively you
could circulate this article as an icebreaker and suggest you would like to
engage in a dialogue.
What are the Right Questions to Ask?
Once you have initiated the discussion you need to know what to talk about.
The exact questions you will need to answer as a group depend on the nature
of your project and the roles of those individuals who are involved. The
following may prove useful as a guideline for resolving common issues.
Does Authorship Matter to All Team Members?
You may find that some team members are not concerned with authorship,
and may not wish to be included as authors. There are a number of
reasons for this; perhaps they are commercial partners who are unable to
accept such credit, or they may have personal reasons for not wanting to
spread their name widely. Not everyone operates in the same reward
economy. Some members of the team may feel they are rewarded amply for
their efforts in software development or web design and may be happy
forgoing more formal recognition. On the other hand, some people will be
keen to show evidence of their outputs and their career development may
benefit enormously from wider exposure. Asking team members if
authorship is important to them is a good starting point for any
discussion.
Authorship Order: Does it Matter?
Keep in mind that not everyone will have the same ideas about authorship
and that their concerns may be grounded in past experiences or real
issues within their own field, such as promotion and tenure. The issue
of authorship order is a good example of disciplinary discrepancy. In
different fields the order of authors has different cultural
connotations. Historians would tend to think the earlier one’s name
appears in a list of authors the larger the contribution that person
made. This idea of “first author” also pervades in fields such as
engineering; however, engineers also reserve the “last author” as the
most senior place, where the principal investigator’s name appears. This
is based on a model in which the principal investigator crafts the idea
for the project and finds the funding to carry it out, but generally a
student or postdoc does the actual research or coding and writing. Other
fields use an alphabetical model for listing authors names, which avoids
the matter of hierarchy altogether. This may be a good solution for some
groups, but one should keep in mind that a reader unaccustomed to
alphabetical listings may try to interpret the level of contribution
each member made based on name-order, regardless of the author’s
intentions.
As each discipline has different ideas on authorship order it is always
helpful to ask your colleagues what if any conventions are used in their
field. Try to look at the situation from everyone’s perspective
individually. Find areas where there are conflicting ideas and those in
which your ideas are the same. By working together you may be able to
compromise on a solution that leaves everyone with a fair attribution
for the work they put into the project. If the issue seems unworkable
because of different disciplinary standards, consider using a footnote
to outline the contributions of each member so that they are able to
highlight their role for future employers. By putting this information
in print, those evaluating job or grant applications have a means of
independently verifying what appears on an applicant’s C.V.
What Level of Contribution is “Enough”?
Everyone likes to feel like their contribution was valued, and this is an
area in which hurt feelings may creep in if the issue is not handled
delicately. Maintain positive and open language and try to keep the
discussion as relaxed as possible. It is a good idea to come into the
discussion with some examples of how similar projects have attributed
authorship in the past. The ICMJE criteria listed above can be a good
starting point for making this decision.
Do not feel pressure to come up with a definitive answer immediately. You
can hold another meeting at a later date once everyone has had a chance
to digest the discussion. But once you have agreed on a way forward, it
is always a good idea to write it down. This puts everyone on the same
page and can be used in the future to resolve any conflicts that arise
over the course of the project. Keep in mind that nothing ever goes
exactly as planned and your group may have to revise the agreement
slightly if certain members fail to fulfill their promises, or if new
members of the team join once the project is underway.
Conclusion
We all need to get credit for our work, whether we’re academics, students,
project managers, or database developers. However, forms of credit vary and
different projects may address the challenge of appropriate forms of citation in
very different ways. FairCite does not seek to impose a new or standard form of
citation. Instead, FairCite seeks to assist in initiating and framing the
conversations between collaborators that can lead to a shared, explicit
understanding of how credit is being assigned and publicized within a given
project. FairCite also suggests that projects provide clear instructions to
authors on how to cite the project in publications, rather than leaving it to
individual discretion. As an appendix to this article, we offer a template with
sample language for a public FairCite declaration reflecting your project's
citation preferences.
Over time, we hope this initiative may lead in three complementary directions.
First, as projects conduct internal conversations and develop expanded forms of
citation, we would like to gather these at the ACH's FairCite page (
http://ach.org/faircite/) and use
them as models for other projects to study and build upon. We encourage projects
to send their models to faircite@digitalhumanities.org. Public discussion may
yield insight into which models prove most transparent, equitable, and scalable.
Second, we hope to see some convergence in practice concerning the location and
discoverability of projects’ FairCite declarations, to make it easier for
authors to locate the preferred form of citation. Finally, with more equitable
and discoverable forms of citation available, we hope to see authors citing
digital projects in their published work and providing visible credit to a wider
range of project contributors.
Acknowledgements
The authors would like to thank Bethany Nowviskie, Stéfan Sinclair, Melissa
Terras, and Ray Siemens for pushing this endeavour forwards. Thanks as well to
Tim Hitchcock for stimulating the idea in the first place. And finally, thanks
to the many people who helped shape FairCite by both offering their ideas and
expressing their concerns.
Appendix: FairCite Declaration of Authorship
We propose the language below as a model or starting point for FairCite
declarations:
The participants named below declare that we have had an open and honest
discussion with all significant contributors about our respective roles in
[project name]
Though our individual contributions took many forms, we agree that these
contributions all constitute an authorship role in the project.
We ask that formal citations of this project use the following citation, to
which we have collectively agreed and which we believe fairly represents the
time, energy, and expertise of our project team members.
[Text of citation]
[Names of project participants]