I use the word "translation" in the way Latour does (2005). The obervations that leads to considering this word particularly apt involve tracing what happens when a statement, and particularly a statement that can do something, travels down some pathway and, at each stop, is transformed to fit the new setting within which "it" will be used. "Translation" is another take of a traditional Boasian insight that, when a object or practice is borrowed from some neighboring population, it will be "adapted" to the local conditions. That is, in more modern language, the original statement (that is the one received from the link above) requires a response sensitive to the new local conditions within which it appears and this involve a "translation" in the usual sense.
For a very apt cartoon that is really a model see what can happen to research findings as the move from a dissertation (Conclusion: A is correlated with B (p=0.56), given C, assuming D and under E conditions) to the response of your "grandmother" who heard something on the news about A causing B.