When a submission gets flagged, a "reason" string is entered by the flagger. This could be done via javascript event binded to the flag link, or by redirecting to a "flag" page that gets the id of the submission being flagged. The flagger then either types a short string, or selects from a limited set of reason from a dropdown box. The reason string is then displayed at the top of a submissions's comment page when the flag is approved as [dead].
The reason string could be added in the title, appended to the byline, as an autosubmitted comment, or as a new heading... whatever is easiest to implement.
Ex. if submission X is [dead] and a user has "show dead" enabled, when they click on submission X it will now say "spam", "duplicate", "inflammatory", "automatic" etc.
This would prevent issues like: http://news.ycombinator.com/item?id=1801727 A post was most likely harmlessly flagged and [dead]ened for being a duplicate of one by a cofounder, but a user who saw it interpreted it as persecution.