I wrote this to explain some things that are often invisible to people who don’t publish in peer-reviewed journals as corresponding authors. My primary purpose for putting up this post is to share that even in the current, imperfect system, there are things that anyone can do to more easily access paywalled papers. Scroll down if you don’t want the preamble.
Background
The cost for a journal article goes to the publisher, not the author. Authors, reviewers and most editors work for free (some editors get small compensations).
Publishing is a very lucrative business. The profits go to publishers, not academics, though academics arguably also benefit to some extent. Because publication is a job requirement, when we publish, we are able to get and keep jobs.
Offering a paper as open access typically costs an academic around 2500–3000 USD in article processing fees. I allocate for these in my grant budgets but sometimes the rules put limits on how much I can allocate to article processing fees, an article may come from an unfunded project, or I may have other uses for those funds that are more important at the time; e.g., paying to do more research or providing funding to a graduate student.
Some funders will also pay the article processing fees for some or all articles, but there may be limits. Many agencies also have policies that require papers to be deposited in an open access repository, but it may take up to a multiple years for them to become available to the public.
Some fields (for example, physics, mathematics, computer science, and economics) have a long tradition of preprints. A preprint is a version of a manuscript that has often not yet by peer-reviewed nor published. I love the preprint model and have used it but preprints are not a full solution to the problem of paywalled articles, for two reasons.
First, preprints have not yet been peer-reviewed. Peer review is not perfect and it will not catch all problems but it does help improve papers. This helps people who are not scientists in that field have some confidence that the conclusions of the paper are valid. This may be especially important for things like randomized controlled trials that can change clinical practice.
Second, preprints are not accepted in all fields nor by all journals. This means that, in some fields, by putting a manuscript out as a preprint, you can no longer submit it to a large number of journals. I think this is terrible and I have personally made a point to talk to editors about allowing manuscripts that have been already disseminated as preprints, but right now this is just a reality of academic science.
How to get copies of articles
Check open repositories
There are many open repositories for article preprints, like arXiv for physics and other mathematical fields and OSF for all kinds of fields. Many university libraries also have open repositories for their own scholars, so checking the university repository of the author might help.
There are also browser extensions that help you in locating openly accessible versions of articles:
You install them into your browser and whenever you encounter a paywalled article, you can try using the extensions to locate an open version of that same article.
Email the authors
As an author, I am thrilled when people want to read my papers. I am allowed to send copies to you upon request and I will happily do so if you just email me to ask. You can usually find an author’s email by searching for them online or, if they are a health researcher, by searching their publications on PubMed and clicking to see the author information.
You’ll have the best shot if the person is the first or last author. This is because some journals require emails from every author; some only require it from the corresponding author.
When you contact an author, you can also ask if they will share any of their other work cited in the paper you’re requesting. This may save multiple emails.
Use your local library
If you have a university, hospital, or public library in your community, you should be able to access at least some articles through them. Not every library has access to every article but they may be able to use interlibrary loan to get access. The more those services get used, the more likely it is that they will keep on existing.
If you spend time in a hospital, especially one affiliated to a medical school, ask if they have a librarian who handles patient inquiries. They may not only be able to get you copies of articles, but also help you find the most relevant articles for your situation.
Check Google Scholar for a pdf
If you put the title of the paper into Google Scholar, you might be able to find it. The right hand side of the search interface will sometimes show you if there is a pdf available. Under the title you may also see a link, “All ## versions.” If you click this, you will see different ways to access the article, which may include pdfs or open access html versions.
Other methods
I do not advocate these methods, especially not as a first line approach. This is because they are not necessarily legal – though they may well be ethical – and perhaps more importantly, I think they take away from libraries at a time when libraries need all the support they can get.
However, for people’s information, other ways to access articles include tweeting a request for an article with the hashtag #icanhazpdf (once someone sends you the pdf, it is good practice to delete your tweet). Another popular approach is to use Sci-Hub, which stores academic articles for direct download. However, it may be blocked by your academic institution. Most attempts at internet blocks can be circumvented by Tor Browser.
Contributing to this guide
If you find something that is wrong on this page just have an idea that should be added, you can easily add it to his guide. Either fork this guide and create a pull request on Github or if you have no idea what that means, just click here and add an Issue where you describe what you want changed and I will make the changes for you.
License
This guide is based on Holly Witteman’s post and is licensed under CC-BY-SA 2.0. The current content is not endorsed by Witteman and has been changed by other contributors.