Any structuralist worth his salt knows full well what he or she owes me. Not to toot my own horn, but if it weren't for all of my signifier/signified stuff, thinkers like Derrida and Lacan would have had to come up with their own buzzwords.
Here are the ideas I think are most important:
The connection between a word and the concept it represents is arbitrary.
I make a big distinction between the system of language and the words that make up the units of that system.
Please don't forget that I am interested in language as it is used at a certain period of time (for example, Japanese in the early fifteenth century) not in a language as it evolved over two hundred years.
I leave it to other linguists to study the way languages change over time. I'm the first to say that language is unstable, that that the signifier and signified are like high school BFFs—meaning that the relationship may be close, but it won't last forever. Yes, the way people use words changes radically over time. Go look up a few curse words in the Oxford English Dictionary if you don't believe me.
People fault me for ignoring grammar and syntax, but words (not sentences) were always my thang.
I reluctantly passed the mantle to Noam Chomsky (he loves sentences) because I knew that if I didn't, he would just steal it and call me a dictator.