Suppose that several individuals are speaking. There are two conversations occurring at once in the same place. Both are heard simultaneously by each person that is present. Each person participates in one and only one conversation.
How does one write dialogue so that it is obvious without explanation that both explicitly quoted speeches occur at once, but without violating conventional grammar?
QUESTION: Specifically, does (1) or (2) as literally heard by the perspective character (PC) violate modern grammar? Which one if any---and in that case, how can it be rewritten without such a violation, yet without rewriting it into (3)?
*COROLLARY**: Are there published modern precedents in novels or short fiction regarding dual but explicit dialogue?
1) There is the em dashery from Tristram Shandy:
"So I told him---This salad is floating---that he doesn't know how---in its dressing!---to fix cars. I look at it and---He needs to get---it's disgusting!---a professional to look at it." Tom spits in the salad and goes on complaining. Janice continues speaking to Bob---who nods and smiles periodically. I don't believe that he can hear what she's saying.
OR
2) There is the symbolic indication of simultaneousness via brackets:
"So I told him that he doesn't know how to fix cars," says Janice. ("This salad is floating in its dressing," Tom complains. "I look at it and---" he spits in the salad.)
"He needs to get a professional to look at it." ("It's disgusting.")
Bob looks at Janice and nods and smiles periodically. I don't believe that he can hear what she's saying.
OR
3) One can summarize who says what and when, or write explicitly the loudest dialogue and summarize what is also said, or separate the speeches and join them with a while clause. For instance
"So I told him that he doesn't know how to fix cars," says Janice. "He needs to get a professional to look at it," she tells Bob.
"This salad is floating in its dressing!" Tom complains while Janice speaks. "I look at it and---" he spits in the salad. "It's disgusting."
Bob looks at Janice and nods and smiles periodically. I don't believe that he can hear what she's saying.
However this cannot achieve the effect which simultaneous dialogues can achieve:
Now I agree: following several overlapping conversations at once is typically confusing if not usually impossible. That's also the point.
"It's noisy and she cannot understand anything that is said ..." is unclear and vague; the reader isn't given the information they need to visualize any realistic group interactions that cause the noisiness mentioned---if that's all the writer tells them. (I am aware that some authors have suggested that many scenes do not need to be capable of being visualized ... )
Several conversations at once that are described as they are heard by the PC is a technical device. It allows the author to reveal information where the PC is present and yet the PC is plausibly unaware of this information despite they fact that they are present there in fact. They hear it but they also have a reason in that context to be insufficiently attentive or incapable of parsing it. (That is not true about the readers, who are made aware of it, because those conversations which are difficult to follow in real time are typically more easily parsed and followed if they are read not heard.)
EDIT:
If I were to identify the speakers explicitly, so that it is not a cacophony like in the case above where the listener is tired and uninterested, I could follow Vollmann's lead, I suppose. But is this really much better?
So I told him---says Janice.
This salad is floating---Tom observes.
That he doesn't know how---Janice continues.
In its dressing! Tom raises his voice, attempting to be heard. "It's disgusting."
To fix cars, concludes Janice. "He needs to see a professional for help."
Meanwhile Tom spits in the salad.
(This question is migrated from english.SE as suggested there.)
In real life, a person will always focus on one speaker at a time. This person may get distracted and switch his attention to another speaker while the first is still talking, but he is aware of this change of focus and it is marked for him by the sound of the voice, and maybe the face of the speaker, so it is never as confusing as unmarked written dialogue, where the reader is given no indication as to who is currently speaking. In reality you will always know which voice you are currently listening to, and if your writing aims to depict this experience you must indicate somehow which voice is speaking which sentences. You can do this by stating that the person listening switches their attention, or by indicating a changed speaker. If you just mix the speeches, as Tristram Shandy does, then that will be an artificial confusion that a real listener would normally not experience.
As for how to style this, normal dialogue conventions are sufficient. If you keep the listener as the subject, write it in one paragraph with the dialogue snippets connected by descriptions of attention shift:
"...," John heard, then another voice caught his attention: "..."
If the speakers are the subjects, give each speaker his own paragraph:
"Bla bla--"
"--bla la bla--"
"--bla."
(I can't find an em dash on my mobile phone, so I'm using two dashes instead.)
Only if you have an unfocussed listener, who lets all the different voices wash over him like meaningless sounds, will you have a true parallel cacophony, but then no voice will stand out and everything will be incomprehensible.
