We have seen how saying that a dream figure is “the same” as one we encountered before is different, both from re-identifying embodied persons in real life and from re-identifying characters in a narrative text.
A problem is beginning to emerge here. Persons in the real world have an objective existence: they (or more precisely, their bodies) have a determinate spatial location and persist through time, they follow a spacetime trajectory; persons in narrative texts are identified by names and definite descriptions, and there are narrative conventions in place that guarantee stable reference. But in dreams, neither applies: dream characters have no bodies with objective existence in space and time, and their occurrences follow no continuity conventions. They’re just fleeting images that can appear and disappear, even shape-shift arbitrarily.
This means that re-identification, if it happens in a dream-journal or by an analyst, cannot be derived from the dream figures itself. Instead, it relies on one of two possible moves: either, the dream figure is first identified with a real-world figure (e.g. the dreamer’s father), and then continuity is derived from that real-world person’s persistence; or else, the re-identification is imported from the dream narration. And the latter is not inherent in the dream, it’s something the dreamer produces afterwards, and in waking state.
And continuity in the dream narrative, too, is not entailed by the underlying dream experience itself. There is no single or “correct” narration for any given dream. In fact, dream narratives might be reformulated in memory multiple times (on waking up, still half-asleep; during writing itself, basing later parts on already written parts rather the dream itself; …) before they are committed to the page. They may be re-written again before an analysis session (or by the analyst, when preparing them for a publication to use them as theoretical example). We never have the “original” or “immediate” dream text, we always just have a product of one or more people’s narrative skills, register and vocabulary, and so on. What’s more important here: we have dream characters which may be unified into a single character only during the narration process. There may be multiple occurrences that are retrospectively identified as the same character by the dreamer. It may happen during writing, or during retelling in analysis. (And then, as an extra step, the analyst might connect different characters with the same archetype, which is a separate unification procedure.) But, to say it again: none of this is dream content. It’s all later application of narrative conventions.
Of course, dream appearances might be so disparate that at some point the conventions don’t longer seem appropriate. A dreamer, when narrating the dream, might start out identifying a character, even re-identifying them throughout an episode, and at one point give up and reach for a narrative device that no longer pretends to realism about figure continuity (“Then the old man transformed into a golden statue which continued to speak…”). It’s natural that a dreamer would try, to some extent, to make narrative sense out of the dream sequence, and likely might try to apply continuity conventions; and there are no hard and fast rules about how far this attempt should go before a dream narration just capitulates and accepts continuity-breaking goings-on. There is, so to speak, an element of narrative style in dream narration: it might depend simply on the taste or playfulness of the dreamer to what lengths they will go in trying to narrate a world of stable, continuous characters or how quickly they will use surrealist devices or fantasy “transformations”, “teleportation”, and the like, in their dream accounts.
The question of when a character is one and the same is not a simple one; and we should be cautious especially when far-reaching conclusions are derived from it, such as “the same archetype is represented here, over multiple episodes, reflecting an ongoing psychological development in the individual”. It may be that it’s not the subject’s dream content that is analyzed here, but their narrative proto-technique (which in most cases, is likely to be picked up from education and cultural influences).
