So far the only argument in favour of should being regarded as a subjunctive form has been along the lines of, “A should-construction has the same meaning as a verb in the subjunctive mood. Therefore should is a subjunctive form.” The flaws in this line of argument should (!) not need to be pointed out, though I have indicated a couple.
Pokemon is not impressed that, “a few authors, Chalker, Greenbaum et al., share your opinion”. Those I chose do know something about English Grammar*. If it were not so tedious, I would produce the opinions of almost every authority on English grammar of the 20th century who share my opinion. The backing of such authorities does not prove my case, but saves me the bother of repeating at length here what they have written
I think the onus is now on those who believe that should is a conditional form to demonstrate convincingly that it is. Cited support from professional linguists would be helpful.
Sidney Greenbaum: Quain Professor of English Language and Literature, University College London 1983-90, Director of the Survey of English Usage, 1983-96, Co-author of A Comprehensive Grammar of the English Language; The Oxford English Grammar.
Frank R Palmer: Professor of Linguistic Science, University of Reading 1965-1984., Author of Grammar, Mood and Modality, A Linguistic Study of the English Verb, Grammar and Meaning.
Sylvia Chalker: Researcher at the Survey of English Usage, University College London. Author of Current English Grammar, Co-author of the Oxford Dictionary of English Grammar, Contributor to the Oxford Companion to the English Language.