I think it should be noted that "re-" can also mean "in reaction/reciprocation"/"back". There are many cases where it's obviously not repeated action.
Examples:
resist: "re-" + "sist" ("stand") = to "stand" strong against some influence
retort: "re-" + "tort" ("twist") = to "twist" an argument back against someone
reply: "re-" + "ply" ("fold") = to "fold" your statement's metaphorical stationery in response to someone else's
request: "re-" + "quest" = to "seek" something from someone
reject: "re-" + "ject" ("throw") = to "throw" an offer back into the giver's metaphorical face
Although I will admit, modern English has leaned away from this, "re-" having been displaced by the adverb "back" to the point where "-back" is practically a suffix. (Think, "Punch buggy, no punch(-)back!") Squirrelous (talk) 12:37, 18 October 2014 (UTC)
- I would also note that the pronunciation of re- is only /ri/ in some words. It seems that /ri/ is used when it is a new coinage, but /rɪ/ and /rɛ/ are common pronunciations in older words. It seems that when re- is attached to a bound root, usually from Latin (whether directly or via French or other Romance language), it usually has the /rɪ/ and /rɛ/ pronunciations. For example, "record" is either /rɪ'kɔrd/ or /'rɛkərd/ and "revolve" has /rɪ/ while "revolution" has /rɛ/. These words also seem likely to be the ones which are less likely to have the repetition meaning. Note that most of the above examples (resist, retort, reply, request) don't have the /ri/ pronunciation. Only "reject" (n) has /ri/ (while as a verb it has /rɪ/). There doesn't seem to be a hard and fast rule though.
- However, the productive morpheme which can be added to any verb, as far as I can tell, only has the meaning of repetition and is always pronounced /ri/.
- As an aside, "back" doesn't seem particularly close to being a suffix. It's possible to say things like "punch me and I'll punch you back" where it's status as a free morpheme is obvious. I also wouldn't describe English using an Anglo-Saxon adverb as displacing a Latinate prefix unless you know it to be a usage of recent coinage. It also seems likely to me that the usages of re- that you're pointing out were imported with the prefix already on the verb. In all the examples you mentioned that is the case, coming from Latin resistere, Latin retortus, French replier, French requeste and Latin rejectus respectively. So they were not the result of English applying the re- prefix productively, but of importing verbs that happened to already have it.