This post has been de-listed
It is no longer included in search results and normal feeds (front page, hot posts, subreddit posts, etc). It remains visible only via the author's post history.
When it comes to words etymologies, excluding words that only are logical connectors and necessary for sentence clarity ie "the", "and", "or", etc., about what proportion of words originate from concrete metaphors?
For example, the words brink, calculus, and suppose all strike me as being reasonably abstract. And yet brink originates within some germanic or nordic language's word for cliff/slope, calculus is derived from the latin word for pebble (as pebbles were used in ancient odometers to assess fees for ancient taxi-like/transportation services), and suppose originates from a word roughly used to say "to put under" hence supposing x puts it under whatever you're about to say.
So what interests me, then, is to know how many (if the word many would be appropriate) words originate from words and thought in and of themselves. One would expect some words to derive from physical circumstances (brink may be abstract, but given the existence of cliffs, it seems logical that the words would correspond) but words like calculus and suppose seem sufficiently abstract that I wouldn't necessarily expect them to have origins that are, essentially, physical circumstances. It would be interesting to me to see just how many words we use to basis our thoughts actually derive from these concrete metaphors of sorts and how many (if any) were actually just thought into existence, so to speak.
Subreddit
Post Details
- Posted
- 12 years ago
- Reddit URL
- View post on reddit.com
- External URL
- reddit.com/r/linguistics...