It [the language] is not so convenient for representing lists of fixed length where one frequently wants the nth element where n is computed rather than obtained by adding 1 to n-1.
I find this is an interesting way of saying it, it made me think for a moment..
Reading this I get an urge to reach back in time and tell the author about all the amazing programs and languages this work will lead to, about Emacs, Scheme, Clojure, ...
For anyone interested in additional context, the paper "AI and the Origins of the Functional Programming Language Style
" by Mark Priestley is a great introduction: https://link.springer.com/article/10.1007/s11023-017-9432-7 (sorry, I don't have a non-paywalled link, but SciHub does the trick).
> implemented, with Nathaniel Rochester, a computer language for list processing within FORTRAN ... the Fortran list processing language (FLPL)
His son David, who would have been 3yo at the time McCarthy wrote this paper, has had his own effect on computer science[2].
[1] https://en.wikipedia.org/wiki/Herbert_Gelernter
[2] https://en.wikipedia.org/wiki/David_Gelernter