If authors must pay, most humanities scholarship will never be open access

There is an assumption here that a commercial publisher needs to be involved, in this case De Gruyter. But across Latin America in particular, universities produced their own OA journals, and don't need a commercial partner. Production on these is done in the humanities and social sciences by academics themselves, sometimes with small institutional support. Aggregator sites like Redalyc and Scielo assure archiving and sometimes Dois across the continent and beyond. By comparison, to show this is possible in the West, I have edited an Anglophone journal out of the US for 20 years in this way, with a Scopus index of over 4, Web of Science, and working with a small team of global North and South scholars. This serves our niche area in political economy/environmental studies well, and our budget is $0. We publish in 3 languages. Worth remembering that the arrival of the internet in 1993 or so changed everything - we don't need commercial publishers and they add little value to outfits like this. Even our journal production software is open source. The world has moved on.

I don’t really understand why the university consortium model isn’t used to found and maintain HSS journals. Still with the state of HSS at UK universities , with departments closing or shrinking daily, research looks to be on its way out anyway.

"For those without funding, who is going to pay for the cost of publication?" Who pays for the subscription to closed access journals? I don't really follow this augment. Institutions are paying the same, irrespective of whether they pay to read or pay to publish.

The main contentious issue is book publishing. Leaving aside the stipulations that UKRI funded research must be published OA now if it is in book form, the REF will set the general policy in this respect. In the humanities and social sciences a very high proportion of submissions to the last REF were books, and OA did not apply then. If commercial publishers are not involved in publishing humanities books, in particular, then that sphere of activity will quickly shrink and eventually disappear, as the disincentives for academics will begin to pile up (because your work, if published in books, will likely no longer be REF-able). But while UKRI are making OA a condition of funding for 'monographs' whether or not this becomes REF2029 policy is uncertain. In 2013/14 I recall being told that all books for the next REF (2020) would have to be OA, and publisher's representatives from the big academic presses - like Taylor & Francis - were going around giving talks in universities and saying it would cost x thousands (at that point over ten thousand UKP, now it will be considerably more) to publish an OA book. I am sure they were rubbing their hands with glee at the thought of really coining it in. The people in charge of research in my institution at the time sent around emails to academics saying 'no more books!' 'Staff writing books will not have their research supported.' This was in the immediate aftermath of the REF 2014 census (late 2013), before the next recycle had even begun. Well, guess what? When the REF census date came around (Dec 2020) they DID accept books, and those same managers no doubt wanted those books that staff had taken to producing often in their own time and unsupported by their institution, to be included in their submissions. Clearly there is a desire to transition to full OA for all academic publication outputs, but until it is adopted as official policy by the REF, it will be in the interests of Humanities and Social Science academics to resist this move. And also to keep in mind that at this stage in the REF cycle, there are always new policies, policy u-turns and adjustments.