Math Is Hard said:
No, doesn't bother me a bit, but I think the inventor of the game might balk. My observation was that all of his trinonyms were true compound words, no separations, no hyphenation.
Oh, right, they were used for a crossword. I guess empty spaces and hyphens aren't so welcome in crosswords.
Still, assuming Merl is reading (
), my point was that if there is any separation, it is only in the writing system, and even then, it is largely arbitrary. There is a 'rule' that we use blank spaces to separate words. But it isn't clear what to do with compounds because, from one perspective, a compound is one word, and from another perspective, a compound is two words. Hyphenation is a compromise, but not everyone uses this or uses it consistently. That's the thing about writing systems: they don't seem to be part of the innate part of language, i.e., there are no innate writing rules as there are other innate linguistc rules. Our writing systems are derived from our speech systems. Throw in all their arbitrariness and it's oftentimes a crapshoot. But I digress/rant.
Anywho, my
satiate example was quick and sloppy, so just forget it. (And that might not even be what happened to
sate.) The basic idea is the same: pieces are added that end up leading back to the original meaning. One thing you might want to look for are what I'll call russian doll http://www.sil.org/linguistics/GlossaryOfLinguisticTerms/WhatIsAStem.htm-http://www.sil.org/linguistics/GlossaryOfLinguisticTerms/WhatIsAnAffixLinguistics.htm pairs, or RDSAP.
The affixes that English mainly uses are prefixes and suffixes. Affixes can be defined by specifyng their form(s), the lexical category to which they affix, the lexical category that they form, and the rough meaning that they contribute (the meanings can be quite peculiar since the meaning of the resulting word is derived from the meaning of the stem). For example,
-er affixes to verbs to form nouns meaning roughly
a thing that [verb]s (e.g., a singer is a thing that sings, a hunter is a thing that hunts).
You want the affix of a RDSAP to have the same resulting lexical category as the stem of the RDSAP. You have some options with their meanings, which are complicated anyway. Obviously, for the more interesting ones, the meanings of the affix and stem would 'agree' in some way, but yah, I'm not sure how to be more specific than that. Boo meanings.
Anywho, the idea is that affix of the RDSAP is the last in the sequence of affixes that lead back to the meaning of the RDSAP's stem. Once you find the pair, you would look for middle steps and hopefully find several options.
Some admittedly uninteresting (you aren't likely to find people actually using them because they are so obviously redundant) but productive ones are affixes that can cancel themselves out. For example, (double) negation can do this with some meanings and doesn't change the category of the stem. So the affixes meaning roughly
not [stem],
un-, in-, the variants of
in- (
im-, il-, ir-), can give you an infinite set of RD words. For example, the set of all stems such that
un-un-[stem] means
[stem] is infinite (
un-un-tie is enough to do that thanks to recursion, but there are probably other roots that work too).
Another uninteresting but promising idea is using stems that include a zero derivation, i.e., where the lexical category is changed without changing the form at all (
to stone / a stone). Then you only need to find one affix (which is presumably easier than finding a sequence of affixes that work together). For the verb
stone, you would want an affix that attached to nouns to form verbs meaning roughly
to kill/bury/destroy with [noun]s. Pretend that
-ate does this. Then you could affix it to the noun
stone to get the verb meaning
to stone,
stonate -- ack! Spelling is so stupid! Lucky thing
-ate has an
e. It wasn't the greatest example anyway.
In addition to, or instead of, looking for stem-affix pairs, you could look for affix tuples, i.e., affixes that you can add in a certian order to get back to the meaning of the original stem. Once you find them, you can look for stems that they work on. Meaning is such a pain though.
It might be easier to find words that match a given affix's meaning rather than looking for affixes that match a given word's meaning since there are fewer affixes than words, i.e., start with affixes and then look for words. But yeah, I'll stop typing now.
ok, that is disqualified for not being a true compound of two words but I thought it was funny.