Ziyad Marar is Global Publishing Director of SAGE and the author of a number of thought-provoking books and articles. His most recent publication is an essay published with the British Academy titled ‘Creating scholarly knowledge in the digital age’ which mounts a passionate and convincing defence of scholarly publishing. On 12 September he’ll be joining a panel of leading lights in the world of scholarly publishing at the ALPSP Conference in Birmingham to debate the point “What is a publisher now?

Ahead of his panel appearance, we caught up with Ziyad to ask him why it’s so important to protect scholarly publishing, the difference between ‘wicked’ and ‘tame’ problems, and what he thought a publisher was in the post-digital age.

JT - In your recently published essay 'Creating scholarly knowledge in the digital age' you defend scholarly publishers. Why do we still need them?

ZM – The first thing to say is that scholarly publishing is not one homogenous block. There are over two thousand publishers ranging from big publicly owned companies, through learned societies, university presses and independents such as SAGE. Moreover, I was focusing in my essay on the value of the publishing process rather than on publishers per se. Whether through journal articles, monographs, chapters in edited collections or textbooks the key feature of the scholarly publishing process is in filtering and certifying that work, so it can take its place in scholarly debate and teaching.

We’re now in a situation where, although information is abundant, human attention is still finite. At a time when data, content and opinion abounds, the question people who consume scholarly content still have to answer is "how do I decide what to read?". Scholarly publishers and journal editors currently play a key role in helping answer that question. This is especially acute in the humanities and social sciences, which make their impact diffusely and over time - often in ways that are somewhat distant from the underlying scholarly or research process - culminating in that final form of expression, whether as a book, essay, chapter or article. In these fields, the publishing process helps to establish knowledge claims as authoritative, which is particularly important to early career authors.

JT - You're also careful to make a distinction between the 'popularity' and 'authority' of a piece of scholarly content. Do you think that culturally speaking we're in danger of mixing up the two?

ZM - I think it is very important to make a clear distinction between the slow, careful work that goes into a piece of scholarship, and the high impact, newsy items that are created with the specific intention of temporarily attracting our attention. By eliding popularity and authority we risk ignoring the fact that these two measurements of success have very different underlying dynamics. We also risk undervaluing the effort required to create authoritative works of scholarship, and the attention and understanding needed to consume them.

What the current model of publishing does very well is provide an incentive structure that ultimately rewards the painstaking, sometimes thankless process of becoming an expert in a particular field. For all their other faults, impact factors do address this distinction in a way that it is difficult to achieve through automated, popularity-based measurements.

While there is a case to be made for the use of automation in certain knowledge fields, particularly certain sciences, the situation becomes much more problematic when we consider how you might automate the evaluation of a piece of research in the humanities or social sciences. It seems to me that much of the enthusiasm around automation emerges from the physical sciences and centres on ‘tame’ problems, to which there are demonstrably right or wrong answers. In the humanities and social sciences, however, research is dominated by ‘wicked’ problems that resist this type of resolution.

JT - You seem to be suggesting that the drive towards automation in publishing risks going too far, and we need to put the human editor back at the centre of the process? Is this true?

ZM - We have to think about what it is that a publisher provides. In the case of a journal editor, for instance, what these people bring is deep and extensive experience in their chosen field. This means they can make nuanced, difficult decisions about the quality of a piece of research that an algorithm simply can’t.

The act of assessing whether one article or a monograph is more worthy of publication than another therefore becomes a complex value judgement best made by someone with a deep knowledge of the subject area and with authority in the wider scholarly community. Fundamentally, good editorial decisions are driven by value judgements, and these are too complex and too important to be left to an algorithm.

To reiterate, this is not to say that existing publishers need to continue, but that the human centred mechanisms of filtering is the heart of what scholarly publishing is about, and needs to continue.

JT - Do you think we would be facing a different vision of the future of publishing if it were social scientists rather than computer scientists and technologists like Clay Shirky who were dominating the conversation?

ZM - I do, and I think it goes back to the way that technology companies develop their products and services. Like certain areas of the sciences, technology is driven forward by ‘tame’ problems – problems to which there can be more or less definitive answers. Over the years, technologists have learned that the fastest way to solve one of these tame problems is to open it up to the ‘wisdom of crowds’. Technology evangelists like Clay Shirky quite rightly operate on the assumption that ‘given enough eyeballs, all bugs are shallow’, but as we’ve seen above, this is an approach that has its limitations when addressing problems that resist analysis and resolution.

There’s a very good reason why humanities and social science researchers don’t write books that offer simple solutions to intractable problems. That’s because important theoretical work tackling wicked, messy problems is based on rigorous care and familiarity with scholarly literature and rarely results in right or wrong answers, law-like generalisations and clear-cut predictions.

Clay Shirky for example is an admirer of Dan Sperber’s book Explaining Culture (published in 1996 by Blackwell). On the wicked question ‘what is culture?’ Shirky says:

‘Dan Sperber has, to my eye, cracked this problem. In a slim, elegant volume of 15 years ago with the modest title Explaining Culture, he outlined a theory of culture as the residue of the epidemic spread of ideas.’

Would that slim, elegant volume have made its way to Shirky’s eye if it had not been published and thus certified by Blackwell? We need to ensure that such intellectual interventions by the next Dan Sperbers to come along are just as, if not more, likely in the future. But in the expanding and accelerating  attention economy, I doubt that Shirky’s own ‘publish, then filter’ formulation will be enough.

JT - Do you think social scientists and researchers in the humanities are better disposed towards the traditional publishing model?

ZM - They are, in general, according to our research. This is just as much a product of the funding environment in which humanities and social science researchers have grown up, as due to the fact that their research addresses much less tractable problems than those addressed by their natural science colleagues. In my experience, those researchers most relaxed about the rise of open access publishing operate in fields where research grants include the cost of submitting that article to open-access journals. It’s common practice for this to be built into research grants from bodies such as The Wellcome Trust. In the humanities and social sciences, however, there are neither the frameworks nor the budgets to allow for this. Most research grants are so small, or more often non-existent, that paid-for submission is unaffordable and researchers therefore depend on the traditional systems to provide a path for publication.

JT - In your essay, you also talked about how the reputation of researchers in Humanities and Social Sciences rests on them having original thoughts. Can publishers do more to protect the IP of these original thoughts, and to do that do they need to look towards trade publishing or into other sectors entirely?

ZM - I think this is a bigger issue for authors than it is for scholarly publishers at the moment. It does, however, raise some interesting questions as to how the vision of knowledge as something that can be measured on an automated basis -which I alluded to earlier- risks turning it into a commodity. The idea that knowledge is something fungible is less troubling in the natural sciences where a large research project can have hundreds or thousands of contributors. But it’s vital that we don’t elide this with the situation in the Humanities and Social Sciences, where knowledge is far less separable from the individual author who created it. This is, it seems to me, why there is a backlash in HSS against the RCUK mandate to publish under a CC BY licence (the most permissive of the creative commons licences permitting unlimited re-use).

JT - Should publishers be partnering with technology companies when publishing their content online or should they grow their technology skill sets?

I think that the wise publisher is agnostic on this issue. There will be times when it is more appropriate to buy in technology or a platform from a company that has demonstrated it can execute it well, and there are times when it makes sense to develop these things in-house. It’s difficult to generalise on this point, but at SAGE we implement both depending on the situation.  

JT - Outside of business models like OA, how important is the question of format to the publishing industry? Should content be book-based, e or print, or on pay wall sites, or in your opinion does it matter?"

ZM - I think the specific format in which individual works of research are published is a secondary question and matters much less than the challenge of establishing new publishing brands for the post-digital era. Let’s not mix up means and ends. As I explained in my essay, it’s entirely possible that current leading publishers and journals will fall by the wayside. What’s important is that there are publishing entities that themselves have the authority and the ability to ‘filter’ submissions in such a way that what they publish enhances the reputation of the author. Whether this takes the form of articles, monographs or the short monograph format that seems to be growing in popularity and importance is not as significant as the assurance that what these brands publish constitutes an authoritative knowledge claim.

JT In fifty words or fewer, what is a publisher?

ZM - By finding, filtering, shaping, curating, certifying, editing, promoting, disseminating and rewarding ideas, a scholarly publisher helps convert them into cultural products that enable their transmission from the minds of writers and editors into the minds of readers.