Holding the house together: how does citing fit with sharing?
Citing and sharing are two of the most important things that academic researchers do for the furtherance of scholarship. Therefore any changes in the way these activities are performed – and in the relationship between the two – are bound to be of interest to everyone within the scholarly ecosystem. And changes are certainly taking place.
Early adopters among the research community have enthusiastically embraced social media. The new requirements of funders are driving institutions in the same direction. Altmetrics and Impact Factor coexist uneasily. The resultant shift in behaviours is putting pressure on the edifice of scholarly communication, and at a foundational level.
The progress of knowledge was always about sharing individual insights and discoveries – but social sharing is too often seen to sit in an antithetical relationship to the traditional citation-driven structure of academic enquiry. Can they work effectively in tandem? And what inferences should publishers draw in their efforts to add value for their customers?
Citing and sharing
Of the two verbs, citing has perhaps the more precise and limited meaning, and (superficially, at least) a more respectable provenance. It is a foundational part of the scholarly skill-set. Citations are the basic building block of how academic arguments are constructed – and also of how academic careers are advanced.
Sharing feels like the new kid on the block because of its association with social media, but arguably has been going on for even longer. It also covers a far wider range of activities. Publishing is a type of sharing. Learned societies, and academic conferences are two others. Correspondence between scientists, for instance, has been of huge importance to scientific discovery, particularly in the early modern period before journals really got going.
According to a piece on the Renaissance Mathematicus blog about the seventeenth century English Mathematician John Collins, a ‘vast interlocking web of correspondence networks’ linked scholars of the time. Letters, often more like scientific papers in their form and the amount of detail they contained, were recopied and sent on to other interested parties in the network, ‘an early modern form of re-blogging or retweeting’.
This peer-to-peer web of correspondence became of less importance with the rise of societies and journals, a centralized and, increasingly, regulated mechanism for scholarly sharing – in part because it was relatively cumbersome and time-consuming, given the available technology of the time; paper, pen and post.
In our own time, however, the affordances of web technology have brought us full circle. Sharing a scientific paper requires only a twitch of the finger muscles, and it can be shared narrowly to one person, or more generally to millions around the world, with equal ease.
Now it is the more centralised means of dissemination that perhaps begin to look cumbersome. To put it crudely, sharing has become a lot more easy than citing.
Wanted: friction-free citation
In a recent post on Scholarly Kitchen, Why Are Publishers and Editors Wasting Time Formatting Citations?, Todd A Carpenter, Executive Director of NISO, bemoaned what a ‘tedious and painful business’ it is writing references; the ‘entirely manual process of checking and formatting’, and the plethora of different citation styles. He called for a cultural shift in publishing practices which would better leverage the developing infrastructure of metadata, DOIs and ORCID identifiers to make the whole thing a bit more point-and-click.
For me this article raises two different but related points that cut to the heart of the changing relationship between citing and sharing.
In the first place, if Carpenter were to get what he wanted (which as comments on the blog point out, might raise some nasty workflow issues) creating a citation would begin to look far more similar to sharing a link.
Citing would, in these circumstances, come to seem more like a subset of sharing – a more rigorous, though not substantially more time-consuming way of pointing to external sources, albeit with a particular, limited purpose.
Secondly, the more similar citing and sharing become, or at least the more similar they come to seem on the surface, the more urgently a need is felt within the scholarly community to differentiate between them, because of one major difference less essential than cultural.
Establishing authoritative sources of knowledge
The shift from peer-to-peer to more centralized forms of knowledge sharing in the Enlightenment – from networks of letter-writing to societies and journal publishing – produced other benefits apart from greater convenience and reach. Chief among these was the introduction of what we might nowadays might call a filter mechanism; a system of scholarly publishing that helped to identify provenance and authority for any particular assertion made within the literature.
This didn’t come all at once. Early editions of the Philosophical Transactions of the Royal Society, the template of modern journal publishing, consisted largely of letters, mostly from correspondents of Henry Oldenburg, the first secretary of the Royal Society (for whom John Collins served as an assistant), which were often of dubious scientific merit.
The authority we now associate with learned journals, the most prestigious of which confer an imprimatur on the articles they publish, was hard-won. When it arrived, however, the value it provided was incalculable. Scholarship became an edifice built with bricks that could be relied upon not to crumble, because peer review and reproducibility had proved each of them sound. Citations were the mortar – or perhaps more aptly, the wall ties – that held the structure together.
But now there is a new way of building knowledge, one that feels by comparison far more frangible, improvisatory and risk-laden. The web was built by academics, and echoes academic working practices in its basic design of pages and links – but if the edifice of scholarly communication is a stuccoed neoclassical mansion, the internet is a shanty town.
No gated compounds on the web
One response of the scholarly community to the flaky quality of the web has to been to reengineer some of its more faulty building parts. URLs are far too mutable, so we have DOIs instead. Search thinks John Helmer is either a deceased haberdasher from Oregon or a journalist specializing in Russian affairs, so we have ORCID to firm up identities. And Google is too undiscriminating for librarians to get their heads around so we have web discovery services.
But while this creation of a parallel, specialised infrastructure for scholarship on the web is hugely valuable, it doesn’t have an answer to the inherent unreliability and instability of entities like Facebook, Twitter and Pinterest and so gives little help in leveraging their virtues. Neither can it provide platforms of comparable utility that are less risk-prone. An entirely robust and risk-free infrastructure for scholarly communication on the web looks unlikely ever to arrive, in fact: there is no credible scenario in which scholarship can benefit from the connective power of the social web without also importing a lot of its risks and messiness.
For this reason, the return of peer-to-peer communication between academics as they embrace social media, threatens to degrade our ability to discriminate, in some eyes. Surely the need to identify authoritative sources of information is even greater on the web than in traditional media (as evidenced by these horticultural hoaxes from Pinterest). So with the best of intentions, academic early-adopters (some believe) are hastening their own demise with all their blogging, tweeting and posting. Sharing starts to look like citing-lite.
The internet, it seems, wants to take us back to the Middle Ages.
Moves to make citing easier, more point-and-click, could similarly be attacked in that they risk opening up the scholarly infrastructure to abuse by so-called predatory publishers (why makes their lives easier) and bad apples among the purveyors of ‘grey literature’ who could achieve greater credibility (and thus commercial gain) by giving their white papers, analyst reports and so on the gloss of academic rigour, without any of the effort this currently involves. If academic authority is something worth defending, surely the barbaroi need to be kept outside the gates.
A false opposition?
This conflict between web-optimists and traditionalists tends to permeate many of the debates in scholarly publishing, including those around metrics. There seems to be a not-always-explicitly-voiced presumption in some quarters that citing and sharing are locked in some sort of death-struggle.
Not so. Principals in the two camps certainly don’t see it that way.
‘Altmetrics is a terrible name,’ Euan Adie, founder of Altmetric, told Research Information at the ALPSP conference recently; ‘it implies a replacement … Citations are always going to be a good indicator of scientific activity. They are a good dataset. In the same way, the JIF is a good indicator too. You just can’t use it exclusively,’
‘The Journal Citation Report (JCR) is known for the IF [impact factor] but in fact there’s a whole range of metrics,’ said Patricia Brennan, VP of research analytics at Thomson Reuters. ‘Metrics and measures are no longer “one size fits all” … What we’re seeing is that there’s a need for an array of metrics and ways to measure the array of activity.’
Not replacement but complementarity
The citing/sharing dynamic is of wider relevance than just metrics however, and in order to understand how citing and sharing fit together in researchers’ behaviour, it is useful to get a feel for what the array of activity that metrics report on might actually comprise.
University websites often carry guides to social media for researchers on their websites. The diagram below comes from the Newcastle University Library guide.
The guide promise to tell researchers how social media can ‘help you at all stages of your research: from exploration to engagement, from planning to publicising, and much more’.
What it indicates is that social sharing has a role to play at many stages of a research workflow, including:
- Investigating a potential area for research
- Keeping tabs on that area as research progresses
- Identifying potential collaborators, audiences, sources of funding
- Polling opinion, crowdsourcing, surveys
- Getting feedback on work in progress (informal peer review)
- Collating and organising links and resources (some of which will eventually become citations)
- Promoting results pre- and post publication
- Assessing/demonstrating impact/outreach
Publishers who wish to develop value-added services for authors (especially those who have now become customers under Gold OA) might see this as a useful checklist. It is also notable that some of the areas on this list are things that librarians might formerly have helped researchers with. Those looking to redefine their roles for changing times could also see a potential advisory and support role for themselves within it.
More indirectly perhaps, for publishers, there is a clear argument here for publisher platforms behaving in ways that support and assist the operation of the social web. This will help them in attracting customers both individual and institutional, and in moving customers along the value chain to become more loyal customers and – through benefiting from their sharing behaviour on social media – advocates.
The very comprehensive array of ways in which social sharing behaviours can support research has shown itself liable to produce fears that change will be so rapid as to threaten traditional standards of academic rigour. Others point to the slow pace of change in academic publishing as compared to the rest of the world and see a greater danger in failing to grasp opportunities in a timely fashion.
All change carries the threat that something important is lost, that babies get thrown out with the bathwater. On the other hand there is no change that doesn’t involve some particle of risk.
Academic discovery, by its very nature, has always involved risk. As researchers reach for a cold beer from the fridge on a Friday evening – after a hard week’s citing and sharing – they might raise a glass to the shade of Francis Bacon, father of modern scholarship, who according to Aubrey’s Brief Lives died of pneumonia contracted while conducting early-stage research into the use of snow for refrigeration.