LanguageLog linking Homestar Runner is perhaps the greatest thing I've seen all day.This is the first I've ever heard of that peeve, too—or rather, I've heard the mispronunciation, but I've never figured it to be anything other than an intentional mangling of the original word, imitating children's speech.
At a certain stage in my life, I had learned the song "Clementine", and I had learned to read the hour hand but not the minute hand on the clock. I decided that when the hour hand was between numbers, it was *clementime*.
In German, it's "Valentinstag". This year, a teacher I know was apparently asked this by one of her pupils why it was called "Valendienstag" (valen-tuesday), even though 2/14 can obviously not be a tuesday every year.
I don't think this is common enough to be a peeve, but I found it rather cute.
As I understand it, it isn't an eggcorn (unless you're assuming that people mistakenly think "valentimes" means "time to valen" which unless they're making some peculiar pun on "valency" seems unlikely). I think it's just that (from my limited, nontechnical understanding) in spoken English it's relatively common for "m" and "n" sounds to blur into each other.
Of course this doesn't answer the linguistic question: Why do people say "valentimes"? It's not like it's assimilation, anticipatory or lag, at a distance or anything. There must be some linguistic explanation.
Maidmc, I'm guessing with "Sam Francisco" something different is going on. My guess is that people aren't actually saying it with a bilabial nasal [m], but, rather, a labio-dental [ɱ]. That is, the quite normal process of articulating a nasal consonant in the same place as the consonant that directly follows. And even though it's written as two words, it's a single morpheme for most English speakers, which may increase the likelihood of that sort of articulation. So, a different process than an intervocalic /n/ becoming /m/.
@maidhc: I agree with Ellen K. Sam Francisco is what I say, more or less, but then I grew up with a version of English in which 'seven' in normal conversation is 'sem.' I think, although I'm not sure, that there's a sort of vestigial glottal stop between the 'e' and the 'm' — it doesn't sound quite the same as 'pro tem,' for example, but I may be fooling myself that there's a difference.
I lived for a while in Sussex, and there's a village called Pevensey that the locals pronounce Pemsey.
The condensation of words is common and most people who do this don't realize they're doing it until it's pointed out to them. And even then, it's hard to accept. It is claimed that Philadelphians pronounce the name of their city as "fluffia". Well, yes, that's one way to represent what it sounds like, but I can't imagine that it's really a regional variation. Anyone (with a reasonably similar accent) would produce that when saying it quickly.
My grandfather, a non-native speaker once asked me why Americans pronounce 20 as "tvonny". I told him that he was mistaken, but then upon thinking about it, "tvon" and "twen" are indistinguishable and, of course, when your native language has no "w" sound, it's natural to hear the version that doesn't have it.
I remember one word from a fifth century Roman schoolboy's list of words to pronounce correctly: "Oculus, not oclus." considering "ojo", "occhio" and "oeil" in the Romance languages, the list did not work so well.
I don't think 'Valentime's' has anything to do with assimilation, as all the other changes in this thread are. Samwich, SaɱFrancisco, etc. are perfectly normal instances of anticipatory nasal assimilation, where the place of articulation of a following (adjacent) consonant spreads onto the nasal. Also as someone noted above, there's no conditioning environment to trigger any assimilation. Dissimilation is also unlikely. So Valentime's looks more like an eggcorn to me, in that it is a plausible, but misanalysed folk etymology.
Febyury is, of course, a result of the same thing that gave us liberry, infastructure and speak. Summbitch is just simple assimilation.
Except "febyury", in addition to losing the /r/, gains a /y/ and loses an /a/. Or did someone forget the A when writing it? While the /r/ changing to /y/ is familiar to me, how I say it, the reduction to 3 syllables is a new one to me.
@Ellen K. Some dialects of UK English, including some high-status ones, reduce syllables in a number of words–Feb-ry, being one example and secretary becoming sek-e-try another. Library can be li-bry, and jewellery (UK spelling) is jew-ler-y. There are also reductions which don't involve dropping syllables which are not common in US English.
@Faldone: Are you sure "Febyury" (AmE "Febuary") isn't the same thing that gave us "suprise" and "govenor", where the first of two [r]s disappears in any environment? I admit "speak" didn't work that way.
@Rodger C: Sorry, couldn't get "Weshchinny". West Someplace?
In German "haben" (in formal speech pronounced [habən], english: "to have" or the first person singular of it) often is pronounced somewhat like [haʔm], where the glottal stop is spoken with the lips closed, in less formal speech and even as [ham] in slang and dialect. The spelling "ham" often is used to indicate dialect or informal speech in literature.
Please excuse my english, for i'm not a native speaker.