As a young teacher, I taught my first senior technical elective course without giving a final exam. I wanted the students to learn, and there were enough assignments. I got feedback from students that without a final, they wouldn’t study, or even read the text. So I give a final now. Not because I want to grade finals, or even use it as evaluation, but to incentivize students to actually read the readings.
This leads me to the hypothesis that the primary purpose of academic Peer Review is not to review papers and give feedback to authors. It is instead to induce authors to submit work of high quality because they believe someone will read it. In practice, the quality will be high enough that the papers will have a chance of passing peer review.
In the absence of peer review, authors could submit any old junk and it would be published. But with review, authors can only submit higher quality pieces with the hope of getting published in a good journal.
Of course peer review does give feedback, this is the developmental component, as well as evaluation (accept, decline) which aids editors in deciding what to accept for publication. I generally dislike the developmental component. It is used to avoid a clear decision of accept or decline, forcing nearly everything into multiple rounds of review, and slowing progress. Further, knowing that almost nothing gets through on the first round, authors don’t submit their best work. I have heard famous faculty say they write papers to 80 or 90% completion, knowing the reviewers will have specific comments that cannot be foreseen, which will finalize the paper. Thus reviewers get incomplete work.
We should restore peer review to its purpose as evaluative.
No paper will ever be perfect, but with the current guarantee that nothing (ok, maybe 1 or 2%) is accepted in round 1, it is pointless to even try. As an editor I am trying to increase those odds, but reviewers do not help in this regard.