Re: I must be going mad!

Because you've just written self-modifying code: that cons is shared with the _function definition_. Watch this: (progn (defun jmh-test-func () (let (tv) (setq tv '((?a) (?b))) (push "testme" (cdr (assoc ?a tv))) tv)) (cl-prettyprint (symbol-function 'jmh-test-func)) (cl-prettyprint (jmh-test-func)) (cl-prettyprint (symbol-function 'jmh-test-func)) (cl-prettyprint (jmh-test-func)) (cl-prettyprint (symbol-function 'jmh-test-func)) (terpri) ) and now this: (progn (defun sjt/safe-to-try-at-home () (let (tv) (setq tv (copy-tree '((?a) (?b)))) (push "testme" (cdr (assoc ?a tv))) tv)) (cl-prettyprint (symbol-function 'sjt/safe-to-try-at-home)) (cl-prettyprint (sjt/safe-to-try-at-home)) (cl-prettyprint (symbol-function 'sjt/safe-to-try-at-home)) (cl-prettyprint (sjt/safe-to-try-at-home)) (cl-prettyprint (symbol-function 'sjt/safe-to-try-at-home)) (terpri) ) NB: it really does have to be `copy-tree', copy-sequence isn't good enough. Simon> It would be good if the implementation triggered an error Simon> for your form though[...]. I don't think so. This is a straightforward, although unintuitive, consequence of setq's reference semantics. -- Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN My nostalgia for Icon makes me forget about any of the bad things. I don't have much nostalgia for Perl, so its faults I remember. Scott Gilbert c.l.py