is granted to a source in exchange for information that could not otherwise be obtained.
The exchange is and the terms of it are almost never explained to the reader.
A reader told that a source is 'close to the negotiations' has been given a phrase, not a fact.
The phrase is with a participant, an assistant and a person who was told afterwards.
Those three sources differ enormously in reliability and the reader cannot distinguish them.
Publications that have tightened their rules require a description of the source's access, not the source's status.
'A person who saw the document' tells a reader something; 'a senior official' tells her nothing she can weigh.
The second is more common because it the source, who has usually asked for it.
A second convention that has spread is to state why anonymity was granted.
Fear of is a reason a reader can evaluate; a wish not to be quoted is a different matter entirely.
Many anonymous sources are not at risk at all and are simply briefing without .
A government official who wishes an idea tested in public can float it anonymously and it by Friday.
The publication has then been used, and the reader has no way of knowing that it was.
Refusing anonymity in such cases costs a story and, over time, costs the official the technique.
Few publications refuse, because the cost is immediate and the benefit is shared with competitors.
This is a collective action problem and has the structure of one, including the usual solutions.
Where several outlets have agreed a common standard, the practice has changed within a year.
Where one outlet has adopted it alone, it has usually abandoned it within two.