We'll survey a novel semantic structure possible in some programming languages, which specifically encodes/represents that several argument inputs had to ultimately derive from a single variable. Structures of this sort have been posited for natural language semantics by Mark Richard, Kit Fine, Angel Pinillos, and others. If such structure is present in natural language, one may naturally expect it to show up in attitude reports, to communicate that a subject is thinking of several arguments as being one. But (if time permits) we'll consider some doubts about whether this structure plays any such dedicated role. We'll explore some other natural language phenomena where this structure may be operating instead.