But does it work in theory? Developing a generative theory for the scholarly commonsPosted: September 2, 2016
The “Scholarly Commons Working Group”
I am part of the Scholarly Commons Working Group at Force11. The goal of this working group is to “define and incubate” a “Scholarly Commons”—something we define as being a set of “principles, best practices, interfaces and standards that should govern the multidirectional flow of scholarly objects through all phases of the research process from conception to dissemination” in any discipline.
As part of this work, we have been working on developing the actual principles that can be said to… well, this is a bit of an issue, actually—govern?, describe (?), organise (?), define (?). Let’s just say, right now, “develop a set of principles that will help in some way identify and establish the Scholarly Commons in some useful, non-trivial fashion.”
The work so far
With the help of a grant from the Helmsley Foundation, we’ve been developing these principles over the course of the last year. In the Spring, we held an invited workshop in which various intellectual leaders within the world of Scholarly Communication and Knowledge Mobilization got together to analyse the problems, lacunae, and disciplinary differences within current publication systems and practices. This group also came up with a list of statements that it felt contributed to the identification, governance, definition, or description of this Commons. In a couple of weeks, we will be holding a second workshop in which another set of leaders in the field will review this work and help further develop the concept.
For the last several months, the Steering Committee of the Working Group has been working on attempting to refine the work of the Madrid participants: taking the principles, observations, and examples produced at that workshop and teasing out a single set of statements around which we think the community might be able to develop a consensus.
The result, in its current form, is a set of 18 Principles group under four main guiding qualities: that the Commons is (or should be) Equitable, Open, Sustainable, and driven by Research Culture. The principles themselves (in their shortest form) are as follows (I should stress that this is a quotation from an in-progress draft that is highly likely to change before it is released):
E1 – The commons is developed and governed by its members through their practice
E2 – The commons is open to all participants who accept its principles
E3 – The commons welcomes and encourages participants of all backgrounds
E4 – The commons assigns credit and responsibility for all contributions without imposing an intrinsic hierarchy
E5 – The commons accepts all contributed objects that adhere to its guidelines on an equal basis regardless of form
E6 – The commons has no intrinsic hierarchies, scores, rankings, or reward systems
O1 – The Commons is open by default: its content and standards are free to read, reuse, and remix by humans and machines, unless there is a compelling reason to restrict access, e.g., personal health information.
O2 – Content is FAIR: Findable, accessible, interoperable and reusable by humans and machines.
O3 – A publisher in the Commons is any entity that will ensure that outputs are open and FAIR.
O4 – All outputs are considered published when they are made available according to the principles and standards of the Commons.
S1 – There is global commitment and participation in the Commons’ long-term viability and preservation.
S2 – All activities and outputs that take place in in the Commons remain in the Commons.
S3 – Use of the Commons cannot devalue the Commons.
S4 – There is an expectation of service by Commoners to support research and scholarship in the Commons.
S5 – The Commons itself is continuously required to respond to the requirements of Commoners.
RC1 – The Commons exists independently of the technology, funding, and business models that support and enable it
RC2 – In the Commons, incentives apply to all stages of the research cycle and are designed to reward behaviours that support the best scholarship
RC3 – In the Commons, the form research is disseminated in is determined by the needs of the research itself rather than the demands of tradition or reward or evaluation systems
Some problems with the Principles as written
I am one of the people responsible both for generating these principles and copy-editing them into their current form. So my criticism of these will therefore hopefully be seen as constructive rather than nihilistic.
But, even as we get them to the stage where they are almost ready for release, I must confess that I am not very happy with them. Not because I don’t agree on the whole with them, but rather because I am not confident that they are that well put together from a formal perspective. Some of the issues that I have with them:
- Why this number? Are there really eighteen principles to the Commons? Or did we just happen to think up eighteen? How do we know we are not missing some?
- Why this distribution? Why are the RC (i.e. “Research Culture”) principles under RC? Do they follow naturally from the idea that the Commons must be driven by research, or are they there because “Research Culture” seemed as good a place as any to put them? Why are the “E” (Equity) principles under E and “O” (Openness) Principles under O?
- What about the mix of scopes and types within the “principles”? “The Commons exists independently of the technology, funding, and business models that support it” (RC1) seems like a different type of claim than “In the Commons, incentives… are designed to reward behaviours that support the best scholarship” (RC2): the first seems like a genuine principle; the second more like good practice.
- Is there not an inconsistency to the kind of entity they define? Sometimes the Commons defined here seems like a consensus among like minded people (e.g. E1); other times, it seems like a club with implicit officers and rules—or at the very least, peer pressure (e.g. S4). When you review the more detailed descriptions we have, you find more and more profound issues like this: there are rules about how one must be identified (perhaps implying some kind of enforcement mechanism); there are rules about what kind of reward systems must be in the Commons and about how there are to be no metrics or evaluations, and so on.
Are they undertheorised?
One explanation, of course, is that we are just sloppy thinkers, or that the principles are still very much a draft. But another explanation—and the one I prefer—is that they are undertheorised. I.e. that our “principles” are not really principles at all, in their current form, but rather a consensus collection of observations, hopes, aspirations, and political goals that have been collected on an ad hoc basis with “little attempt to analyse, explain, draw out common features across situations, identify patterns of behaviour, syndromes of factors, and so forth” (See Woods 2006 ‘undertheorized’). The problem, in other words, is that these have been thought up (or observed) on an ad hoc basis, rather than generated from any underlying theory of how the Commons works or ought to work.
There is nothing to be ashamed of in saying this: any project like the Scholarly Commons project needs to bounce back and forth between observation and theory. You can’t, or at least probably shouldn’t, theorise without gathering any material for your theory to explain; and, conversely, you can’t, or at least probably shouldn’t, continue to collect ad hoc examples without stopping every so often to check if these examples do suggest some larger theory that accounts for them. You know you’ve reached the end of your work when your theory can be used to generate from first principles claims that look very much like those you originally developed by observation; or, conversely, that you can map the observations you made onto an explanatory set of principles (I realise that some scientists might object to this as ‘HARKing’—Hypothesis After Results Known. Since I’m a Humanist and defining principles like this is a Humanities question, however, I don’t have any problem with this: in the Humanities Harking is such a basic methodology, you’d think we all have a smoker’s cough)
So, applying all this to our observations of the Commons, is there in fact a real set of principled principles—that is to say, a set of statements from which everything in our observations can be derived or implied?
In fact I think there is. In anticipation of our upcoming workshop in San Diego, here’s my attempt to develop a set of principles that describe what participants in the workshops thus far have meant when they have discussed “the Commons”:
P. The Scholarly Commons is a consensus among knowledge producers and users that
P1. research and knowledge should be freely available to all who wish to use or reuse it;
P2. participation in the production and use of knowledge should be open to all who wish to participate;
P3. our practices should be such that there are no systemic barriers and disincentives to prevent either free use or open participation.
R. On the basis of these three principles there are four basic rules to the commons:
R1. Participation and access are the only intrinsic reward systems within the Commons. The Commons does not itself have systems for rewarding participation in any other way1;
R2. The Commons does not require the use of any specific technology, approach, process, or system2;
R3. The Commons does not prevent the development of either external systems for either reward or specific technologies, processes approaches, and systems, but such rewards, technologies, processes, approaches, and systems cannot be part of the definition of the Commons3;
R4. Commoners may not participate in external activities that hurt the viability of the commons.4
How do they line up?
With the exception of the last rule (R4), which I am not entirely sure is a core value, and the third principle (P3), which might just be a negative restatement and operationalisation of the first two, I believe that these principles and rules can be used to generate all non-contradictory principles put together by the Scholarly Commons Working Group participants—and where they can’t, the fact that they cannot is actually an indication that there a problem with the original principle: that it either contradicts something else, is not actually a core value, is formulated poorly, or has some other problem.
To demonstrate this, here’s a table, in which I map each of the new principles against the original 18:
|New Principle||Original Principle(s)|
|P||E1, E2, O3, S1, S3, S5|
|P1||E3, O1, O2, O4, S2, RC1|
|P2||E4, O4, RC1|
|P3||E4, E5, E6, O2, O4, S2, S5, RC1, RC3|
|R1||E4, E6, O4, RC3|
|R2||E4, E5, E6, O4, RC3|
|R4||E2, S1, S2, S3, S4|
This suggests that the only original “principle” that can’t be reconciled with my more compact formulation, is RC2 (“In the Commons, incentives apply to all stages of the research cycle and are designed to reward behaviours that support the best scholarship”), and that there is one new rule that doesn’t map easily onto any of the original ones (R3 “The Commons does not prevent the development of either external systems for either reward or specific technologies, processes, and systems; it is just that such systems and technologies cannot be part of the definition of the Commons”).
I would argue, however, that these are related to each other and that neither is a fatal flaw. In the case of RC2, the original principle seems to me to be inconsistent in its current formulation with the rest of the original principles: incentives, by their very nature, discriminate against non-incentivised activities (in this case, non-“best” scholarship). This could be understood as violating original principles E2, E4, O4, and especially RC1 and RC4 incentives that are “designed,” moreover, imply a central governance and, perhaps, enforcement mechanism that is inconsistent with original Principles E1 and E6. (Written negatively, on other hand—i.e. “In the Commons, incentives do not reward behaviours that harm the production of the best scholarship”—principle RC2 in fact maps onto new principles P3, “R1”#r1, and R4).
In the case of R3, I think that the problem is a related lacuna in the original text: there is actually a lot of discussion in the description and operationalisation of the original principles about how and to what extent the Commons is compatible with existing systems and participants, such as major commercial publishers and university promotion and reward systems; but with the exception of the problematic RC2, there is no specific principle in our original set discussing, in abstract terms, how the commons interacts with these structures, systems, and participants. Taken together, I believer that the new principles expose (and clarify) some issues with the question of rewards inherent to the old statement.
In a couple of weeks we’ll be presenting some version of the original principles to our workshop participants for comment, criticism, and revisions. While I’m sure that the proposal here must be missing something, I’m putting this up as a blog now so we can use it in our preparations and discussions for that workshop.
1Anything else would potentially run foul of P3, since there is no such thing as a perfect reward system and reward and incentive systems work by privileging certain kinds of work or behaviour and, as a result, creating systemic barriers to others.
3This is because constructing a system that prevented the development of external reward systems or proprietary technology would violate all three of the core values: it would systemically disincentivise access and participation (i.e. by those, such as commercial presses or Universities, who wanted or needed to design such systems.
4This seems to be required, or else the Commons would rapidly cease to exist. But I’m not sure it actually belongs here, because I can’t see how an enforcement mechanism is consistent with literally the first principle: P.