Library access is one of the most important problems that the independent scholar needs to solve. Fortunately, it is also one of the most tractable.
Traditionally, one of the perks of a university appointment was access to the university library system. In the pre-internet days, this access was an absolute necessity. Today, however, most journal articles are published electronically, and much of the back catalog has been digitized, as have many books. That means that, in principle, all of that information can be accessed anywhere, by anyone.
But, while there is a growing movement toward open-access publication, where published research is made freely available to the public, many journals, including some of the most prestigious, still make their money by charging readers for access. At prices of $25 or more per article, purchasing access as an individual quickly becomes prohibitively expensive. Fortunately, there are a number of work-arounds that, with a little bit of time and effort, should allow you to access pretty much everything you need.
1) Courtesy Appointments. If you have connections with a university — maybe geographically close to you — you may be able to get a courtesy appointment or affiliation with them that would come with login credentials. If you have time to do a bit of adjunct teaching, this might be a way to establish such a connection. While adjunct pay is so poor that it may not be worth your time for the paycheck, if teaching, say, a course a year gives you library access.
2) Alumni Access. Many universities have programs that give alumni partial or full access to library resources. For example, JSTOR has a list of over 100 colleges and universities that offer alumni access to their stuff (http://about.jstor.org/alumni#Institutions-in-program). Some universities give alumni full-fledged access to library resources online.
So, check with your undergraduate and graduate institution(s) to see what sorts of possibilities exist.
3) Public Libraries and Local Universities. For certain resources, public libraries may have subscriptions, and some universities, particularly public ones, have a mandate to allow public access to library resources.
But even some private schools permit library access by the general public. Typically, these options require you to physically go to the library. But some may be willing to hand out log-in credentials — it never hurts to ask!
4) Google-fu. In my experience, more than half of the paywalled papers that I want to access are actually available for download somewhere on the web. A lot of researchers post PDFs on their personal or professional websites, and/or at places like ResearchGate and Academia.edu.
Here’s what I do: from Google Scholar, if I see an article I’m interested in, I click on the “All X versions” link. Sometimes only the publisher’s link shows up on the main search page, but you can find one or more self-archived versions — wth links direct to a PDF of the article.
5) Unpaywall. If googling the thing you’re looking for seems like too much work, you can install the unpaywall browser extension (for Chrome and Firefox), available here. When you go to an article page on the journal’s website, this extension will tell you if there is a freely available copy somewhere, including the author’s home page, Academia.edu, ResearchGate, and the arXiv.
6) Preprint servers. In many areas of the physical and mathematical sciences, it has long been the norm to make a copy of your work publicly available on the arXiv preprint server before it has been published in a peer-reviewed journal — perhaps even before it has been submitted. In recent years, analogous preprint servers have been established for a number of different fields.
Ideally, a preprint provides an early view of the most recent work. Then, after the paper has gone through the standard peer-review process, the authors update the preprint to reflect any changes made as a result of peer review. Then the preprint has all the same information as the final published work, lacking only the journal’s layout. However, if you’re using a preprint version of a published paper, check to make sure you’re getting the most recent version.
In many, but not all, cases, search engines like Google Scholar recognize preprints and final publications as versions of the same article. However, it might be worth bookmarking the preprint server(s) that people use in your field.
In addition to the original arXiv (http://arXiv.org), here are some others:
bioRxiv (http://biorxiv.org) — biology
PeerJ Preprints (http://peerj.com/preprints/) — biology
CogPrints (http://cogprints.org) — cognition, including psychology, neuroscience, linguistics, computer science, philosophy, anthropology, etc.
You can find many, many more repositories, including a lot of institution-specific repositories, by searching on the Directory of Open Access Repositories (http://www.opendoar.org).
If there’s an important preprint server that a lot of people in your field use, let us know.
7) Contacting the Author. Even if you can’t access a given paper, you can access the contact information for the corresponding author. Shoot them an e-mail and ask if they would be willing to send you a PDF copy of the paper. Most people are happy to do this. In fact, many will be flattered by the fact that you were interested enough to ask. You don’t need to make excuses or justify the request. If they’re an old, like me, they’ll remember when it was commonplace to mail out paper reprints in response to requests (which came on “postcards” through the “mail”). If they’re young, they’ll be delirious over the idea that someone other than them is going to read their paper.
8) Sci-Hub. Sci-Hub allows you to download just about any academic paper. We can’t necessarily endorse this method, which is not legal in most localities. So, purely for informational purposes, Sci-Hub can be accessed at http://sci-hub.bz, http://sci-hub.cc, or http://sci-hub.ac.