<<What about those false friends though? The immediate one that comes to mind is 'Gift', which actually means 'poison' in German (das Geschenk being the word for a gift/present).
Does anyone know whether this is mere coincidence, or whether there is actually some connection. Maybe someone in the olden days presenting a member of Royalty, who they wished to dispense of, with some sumptious food, which was actually laced with arsenic...>>
Hehe, yeah that's an interesting one. English "gift" and German "Gift" are etymologically related but the semantic difference is indeed striking! I remember reading an explanation about how it happened but I can't find it right now. Interestingly, this topic just popped up on a linguistics forum I often read and while it hasn't garnered that many responses yet as of me writing this you may want to check back when some more people have responded:
http://www.spinnoff.com/zbb/viewtopic.php?t=17092&highlight=gift+german+poison
Semantic drift can be a very interesting thing. A famous example in German is the shift from Old High German "thiorna" (virgin) → Modern German "Dirne" (whore). Sometimes you give a language a few centuries and a handfiul of its words can come to mean exactly the opposite!
<<Oh yes, and I as I have written 'olden' which seems right in this context, as opposed to old, does anyone know where this form originates from. I assume it is a leftover of an earlier form of English. Is it a relic of when English was a more highly-inflected language though? >>
Yeah I think it's a relic from a more inflected period in the language. Oed seems to agree--this is what oed.com has to say on the matter:
"Phrases similar to olden days were common in both Old and Middle English. Old English had the phrase (on) ealdum dagum (cf. OLD a. 9a), with dative plural ending -um. In Middle English, such phrases often preserved archaic forms. For example, from early Middle English onwards the phrase (bi) olde dawe is attested, where the noun has an archaic plural form (cf. DAY n. A. ), but the adjective has the regular marked -e form. In late Middle English (bi) olden dawes, (in) holdon daw (see quots. a1400 and c1426), archaic forms of the plural noun persist (day usu. having a regular plural form elsewhere in both texts), suggesting that the phrase had become a fixed formula. Their -en element could conceivably be the reflex of the Old English dative plural ending -um, accidentally not attested during the intervening period (although -en is very occasionally found as a plural inflection of this word in other senses in early Middle English); but an alternative explanation, possibly more consistent with the non-attestation of -en forms earlier, would be that the trisyllabic structure of the fixed phrasal unit was preserved by means of the replacement of unstressed -e (normally lost in this position) by -en (protected in medial position before a homorganic consonant). From fixed phrases like in olden days the word was perh. generalized in early modern English to other contexts.]"
<<An interesting discussion, I was interested to see that 'failure' is not actually related to 'Fehler', I had always assumed it was.>>
Yeah that's another interesting thing is that it's common for any two languages (whether related or not) to have a certain number of seemingly corresponding words yet it's just by coincidence (often if you look at older forms of the languages the forms are less like each other so it's just been by random sound changes that the modern forms resemble one another). An example of this is the words in Germanic languages for "have," which superficially look like Latin "habere." However, thru normal sound changes from Proto Indo European the Germanic words for "have" (English "have" Swedish "ha(va)" German "haben" etc.) are actually related to Latin "capere," both coming from Proto Indo European *kap-. As it turns out Latin "habere" is actually related to the Germanic words for "give" ultimately from Proto Indo European *ghabh-.