Librarians at one North American university have posted on the internet a guide to help students distinguish a scholarly journal from a popular one. Among the telltale signs of populist offerings are high production values - "slick, glossy, with colour pictures, photographs, and illustrations" - and writing that is "non-technical" and uses "simple vocabulary accessible to the majority of readers"; whereas textual material in a journal is described as "college level, in the specialized vocabulary of the discipline covered".

That's akin to saying that scholarly articles are written in code for those in the know; they are more exclusionary than inclusionary. There is a widespread belief that any work that is easily understood by a non-specialist must have been dumbed down. References and footnotes therefore swaddle the text like an intellectual security blanket and deter the curious reader.

Of course, academic writing must have rigour, but not so much that it "suffers from rigor mortis", as John Cornwell, director of the Science and Human Dimension Project based at Jesus College, Cambridge, points out in our cover story. Andrew Franklin at Profile Books agrees. He publishes trade books, aimed "at people who want to read books rather than have to read them".

That remark gets to the nub of the problem. Peer review, the research assessment exercise and even book reviews do little to encourage accessibility. If scholars are read and judged only by their peers, that is who they will address. Authors of "popular" books that shift more than a hundred or so copies are accused of selling out. But shouldn't disseminating the wonderful fruits of academia to the masses be something to strive for, or is there a secret prize for the scholarly work that can be understood by the fewest readers?

É«ºÐÖ±²¥

For some, the internet, by making many primary sources readily accessible, is seen as a threat to the specialists and their sense of uniqueness. But it can also be seen as a huge opportunity to reach out to the public.

Academics in the digital humanities are doing just this. University College London will next month use "crowdsourcing" in its "Transcribe Bentham" initiative. Amateur historians, schoolchildren and others will help read and transcribe 10,000 folios of Jeremy Bentham's handwritten documents.

É«ºÐÖ±²¥

This project may allow scholars to interact with thousands of non-specialists in its execution, but the results will still be processed by the stultifying academic mill. Melissa Terras, deputy director of the UCL Centre for Digital Humanities, writes in her blog: "... no matter how successful Transcribe Bentham, the 'impact' will be felt in the same usual way - through publications. This is a nonsense, but it's part of the academic game ... It's not enough to make something that is successful and interesting and well used: you have to write a paper about it that gets published in the Journal of Successful Academic Stuff ... ".

As research becomes more interdisciplinary, requiring work in many fields to be comprehensible to non-specialists, and as calls for evidence-based policymaking grow, "non-technical" and "simple vocabulary accessible to the majority of readers" cannot remain antithetical to scholarly writing. The public should be able to access and understand the work that they have in many cases funded. An overhaul of the academic publishing system is long overdue.

ann.mroz@tsleducation.com.