Fortran was being devised at the same time that Chomsky, et.
al. were doing the research on formal languages. Essentially, Fortran
syntax was devised before computer scientists knew better. I suspect
COBOL has similar issues.
I agree. I think you nailed about "not knowing any better.' And it really is amazing how well it has endured. Like a lot of things, the first guys do make errors because they really have not yet eccounter the longer term issues - they are just trying to solve the problem they had in front of them. On another fron, just think the issues in networking that were made in the 70s. Same thing. But IP/TCP works and works really well and has endured.
You folks in the compiler team and in particular the Fortran crew, did a great job over the years. I used to kid your boss asking of there were Fortran developers than customers in our DEC days; but all kidding aside. And he knew I knew. I have always respected those folks. As I said, they paid my salary for so many years.
I think it's sad we don't teach Fortran to 'modern' programmers in a comparitive languages class. The students should know what is good, why it has lasted and marvel at what a wonder system people devised in the late 1950s and how well Computer Scientists have over the next 60 years kept it strong and relevant. Then you can teach them, Rust, Go or whatever the cool kids think are hot and important. Ask them all, why do we think this languages will or will not last (and the end, it will be economics but that's another thread).
Clem