It certainly is interesting to see how things have evolved. Back then, phrases were not used as answers. Very few fill-in-the-blanks (I think I saw two or three in all of these MJ puzzles). Many clues start with "A", "An" and sometimes "The". I think the first two may have been used to differentiate between verbs and nouns (or adjectives), but not always. And "The" seems to appear arbitrarily, such as "The pineapple" for ANANA. A lot of obscurity, that's for sure. And interesting cluing.
I'll say, phrases were used as answers back then, but very rarely. I'd have to really dig in the stuff I have to find an example, but indeed it's nowhere near common.
As for "A" "An" and "The", there's a lot more attention to grammar and word meaning then than now. I remember talking to a person that temped in a school and ended up getting to grade some essays. Most of them ended up with F grades based on grammar issues, incorrect word usages, misspellings, and the like. He ended up getting reprimanded for grading like that and was told to give rosier things. I've even noticed my older professors when I was in college would eviscerate my papers for those kind of issues. I even had one that suggested the College Writing Lab as a proofreader get egg on his face when I mentioned to him that said group passed it with flying colors and said I "write wonderfully". (He ultimately allowed us to resubmit.) That said, I find I could red pen a lot of newspaper articles I read for those kind of things (the biggest strike being over-reliance on spell check) and get very frustrated at times with how badly word usage gets mangled in these modern crosswords. I know educational standards have gone downhill greatly, even since I went to school.
That said, what I'm finding with these old crosswords is there is a demonstration of how different English usage is over time. I think it has changed over my time on this Earth, even. Of course, you can notice it in a stark way reading Elizabethan language. I do know a lot of the "classic novels" get edited from the originally penned version when they get reprinted for these kind of reasons. My question I have on all of these things is how solvers of that time would have perceived them.
What I'm noticing in doing old crosswords overall as to where I have problems:
1. Novel words. Things I've never heard of. I do have dictionary ads in those puzzle books I mentioned. Along with other references I've read of people hogging use of dictionaries in the libraries, I have to think a certain amount of dictionary trawling was expected, since the constructors likely did the same to fill in a grid. (Advantage software there.) Your word usage listings will definitely highlight those spots for me, though I'm surprised some of the words that I do know have been sparsely used or not at all.
2. Words I know used in novel ways. Usually I get enough crosses and guess something thinking "I've never heard of it, but that makes sense." I figure this is a mark of how the language has changed.
3. Overly vague clues. I have trouble with this in crosswords, period. But I notice it to be more pronounced in these older ones. Again, the clue words maybe had more meaning to solvers then than now?
As a side note, I discovered what appears to be a few typos in the clues of those PUZ files. Nothing hardly taken as I try my best and still notice them when I transcribe crosswords myself and go back to solve them later once I've forgotten any answers I might have transcribed when making the PUZ file.