I’ve long wondered how in reported speech, what sort of change in nuance is produced by switching around the normal order of the subject (that is, the speaker) and the “speech-related” verb (such as say, ask, or mutter) used for simple declaration of whatever was said.
Assuming that QQQ below stands for any quoted speech, what — if any — subtle shifts of meaning are there between the following three formats?
- He said, “QQQ.”
- “QQQ,” he said.
- “QQQ,” said he.
In particular, in what situations is the “inverted” VS1 order of said he preferred over either or both of the two versions that use he said in the “normal” SV2 order?
Are these nothing more than three equal options that vary by individual writers’ personal tastes?
If not, then what rules exist for choosing between and distinguishing these three variants?
1. VS means verb–subject, or with object, VSO or OVS.
2. SV means subject–verb, or with object, SVO or OSV.