First I will echo @julianstirling and @unixjazz’s responses: SocArXiv, Zenodo, and the Open Science Framework (OSF) are great for publishing research output including the social sciences and humanities (though I admit I sometimes I have trouble differentiating the two, sorry!).
Since you mentioned “essays”, I wasn’t sure whether that is intended to be “peer-reviewed papers” in an academic journal? In case the essays don’t fit the rigid mold of academic journals, may I also submit for your consideration the journal Research Ideas and Outcomes (RIO)? It is designed to allow publication of any research output, from documents, presentations, grant proposals, media, to many other things using an open license. Granted, Zenodo and OSF can also publish different types of research output, it wouldn’t hurt to take a look at RIO, too.
This friend is even considering creating a publication for a groups of researchers she works with.
That’s exciting! Every time something like this comes up I must give my obligatory, strong suggestion: Please don’t choose Elsevier as the publisher! They actively undermine open science and open access efforts worldwide. In contrast, Pensoft, the publisher of RIO, and Ubiquity Press are big advocates of open access journals, but as far as I know there isn’t an open source journal publishing platform (is there?).
How familiar is your friend regarding licenses? It’s important to carefully think through which license to use (e.g. Creative Commons Attribution-ShareAlike 4.0 (CC BY-SA 4.0)) and how to correctly apply it before publishing.
Hope this helps.