The Half-life of Euphemisms
For no particular reason, I was recently thinking about the futility of the euphemism strategy—replacing a word that has negative connotations in the hope that the change will get rid of the connotations. The problem is that if, as is usually the case, the connotations are based on what the word means not how it sounds, they will rapidly transfer to the substitute. The record for sequence length may be held by what we now usually refer to as a toilet. I do not know what the earliest term was, but the string includes "privy," "guarderobe," "WC," "lavatory," "bathroom," "toilet," and probably more that I have missed.
A different example that I noticed a few years ago was "gay." It was introduced as a substitute for "homosexual" on the theory that the latter was an insulting term. The problem, as usual, was that what made it insulting was that many people regarded what it described as immoral, disgusting, or both—and although such feelings may weaken over time, they are not eliminated by a change in label.
Not only did the negative connotations spread to the new word, the effect was not limited to its euphemistic use—a fact I discovered listening to casual chatter on World of Warcraft. Posters routinely used "gay" as a general purpose negative term, often with no connection to homosexuality. "That's gay" meant, more or less, "isn't that terrible."
There used to be a theory known as the Sapir-Whorf hypothesis, according to which language shaped thought. As best I can tell, the supposed evidence for it was mostly bogus. The euphemism strategy is the applied version.
And doesn't work.