Almost 50% of scholarly articles are now open access in some form. This greatly benefits scholars at most institutions and is especially helpful to independent scholars and those without access to libraries. It also furthers the long-standing idea of knowledge as a public good. The changing dynamics of open access (OA) threaten this positive development by solidifying the pay-to-publish OA model which further marginalizes peripheral scholars and incentivizes the development of substandard and predatory journals. Causal loop diagrams (CLDs) are used to illustrate these interactions. THE RISE OF OPEN ACCESS PUBLISHING The Generation of Academic Knowledge Much scholarly work is built on knowledge discovered or created by previous scholars. The details of that previous work are communicated via scholarly publications. Although the form of these has changed over the years, the most common form now is via journal articles and books. Access to this accumulated knowledge is an issue of vital importance to scholars Volume 9, General Issue JL SC 2 | eP2345 Journal of Librarianship and Scholarly Communication around the world because, until recently, much of it was not available to most of them. Although open access (OA) publishing has made accumulated knowledge more available, scholars must also be able to publish their own findings so that others can benefit (Figure 1). This article examines how the evolving open access movement is changing the dynamics of scholarly publishing in a way that both helps, but also hinders, peripheral scholars. Causal loop diagrams are used to illustrate factors that cause, and reinforce, these trends, sometimes making them difficult to alter. 2 A Note on Reading Causal Loop Diagrams. Causal loop diagrams are diagrammatic representations of links among elements in a system: in this case the system of interlinked factors that affect open access publishing. Typically, elements in a causal loop diagram are connected with unidirectional arrows which link a cause to an effect. These arrows, or causal links, indicate how a change in the causal variable might affect change in the second variable. These causal links are generally thought of as causing either: 1. Change in the same