Consider the program at the bottom of page 29 that calculates the length of a list. What's the delta to get it to run in X? In Clojure, quite a lot of modification is needed. Scheme, slightly less. Common Lisp though, all you have to do is change the archaic top-level "DEFINE (( (LENGTH ..." to "(defparameter LENGTH...", and "ADD1" to "1+", and to call it change "LENGTH ((A B C D))" to "(LENGTH '(A B C D))", similar for "LENGTH (((X . Y) A CAR (N B) (X Y Z)))". The features of cons cell lists (including the syntax (X . Y)), lambdas, PROG, setq, labeled gotos, and even being able to use SHOUTING names if you want, have all been retained throughout the decades.
The program from 48-51 works as an example too. Here's a small driver someone wrote in CL to run it without having to modify it: http://www.informatimago.com/develop/lisp/com/informatimago/... One notable feature is that of being able to TRACE a function, which is quite handy, and is retained in modern Lisp systems. You could write such a driver in any language, sure, but it wouldn't be as small and simple in anything but Common Lisp, because so many of the functions in Appendix A have been carried forward through history with many Lisps until Common Lisp, so you can just pass them through to the base Lisp, no need to remap and wrap things/write a small interpreter like you would in Clojure/Scheme. A language that just ignores all of that history and makes no attempt to be backwards compatible with any of it can't be considered a "true Lisp", even if it shares other Lispy features.
But again, it's a tired argument, and whether someone disagrees isn't very important. To me, Clojure is a Lisp-inspired language (like so many others, e.g. Julia), and is attractive on its own merits, without having to crib the good name of Lisp.
There are purists who don't consider Racket or even Scheme to be Lisps, either. FWIW, I do not agree with them (I also consider Clojure a Lisp), but they exist.
There's a certain highly vocal contingent of Lisp fans who consider the Common Lisp spec to be some kind of sacred document, and that anything that's been invented in the past 37 (!) years is useless.
IMO, those people have done far more harm than good for Lisp and Lisp-like languages.
And yeah, there's a lot of the same religion-like purity around Rust nowadays.