Voice and viewpoint are inseparable, no matter what viewpoint the writer is using. This is true of all viewpoints to some extent, but it is most evident in first person.
In first person, the viewpoint character is the putative storyteller, so that character’s voice is the narrative voice. But in everything other than fiction, an author writing “I went to New York on my summer vacation” is telling the reader about what happened to the actual writer on their real-life summer vacation. In fiction, the actual writer may really have gone to California on vacation, or not taken a vacation at all. It’s the fictional character who is narrating the story who went to New York; the writer may never have been to that city at all.
Because we are all accustomed to writing first-person school essays and twitter posts and emails and Instagram labels about what we are doing or have done, it can be extremely difficult for a writer to assume the character and voice of a fictional person who is not them. There is a lot of mental inertia to be overcome before one can actually write as if one were someone else. (This is why one of the often-prescribed exercises for budding actors is to write letters from the character they will be portraying to some other character in the play or film.)
There are two approaches to this problem. The first is to write a character who is close to autobiographical, in terms of personality and voice; the second is to write a character who is as completely different from oneself as possible in terms of everything from height, age, and appearance to life circumstances and morality. Either can work, or not.
The self-insertion character can become either a Mary Sue – a super-idealized, super-successful version of the author that nobody will accept as realistic – or a totally believable Everyman with whom readers can easily identify. The completely different character can be either an implausible stack of stereotypes that display the writer’s utter lack of experience with the chosen differences, or an empathetic look at the things human beings have in common underneath all the obvious differences.
Which direction the writer chooses to go – and how successful they are likely to be at depicting their narrator-of-choice – depends to some extent on how well they can balance the conflict between the real-life “I” who is the writer and the fictional “I” who is the character. Conflating the two can result in treating the fictional “I” as if it were the writer…and since nobody wants to write themselves as deeply flawed or going through a series of painful failures, the fictional “I” slides closer to the too-perfect-to-be-real Mary Sue. This writer may be better off working with a fictional “I” who is completely different, even though this means the writer will have to pay constant attention to the character’s voice (and personality, and moral decisions) to keep them all in character.
By the same token, a writer who is very private may have difficulty in revealing a too-similar viewpoint character’s thoughts and emotions, because it seems too much like telling the world their own, very personal, secrets. This writer may find it easier to start with a first-person viewpoint character who is very, very different from themselves, so that it is easy to remember that this is a fictional “I” and not a thinly-disguised autobiography.
The intimacy that is implied by first-person affects readers as well as writers. Having “I” tell the story invites the reader to identify with the narrator and sympathize with their trials and tribulations. Most readers, however, don’t want to find themselves sympathizing or identifying closely with a narrator who repels them; that Nabokov’s Lolita succeeds in doing this (at least for a lot of readers) is why it’s considered a work of genius.
A writer who wants to write an obnoxious, inconsiderate, or downright evil first-person narrator thus has an uphill battle when it comes to holding on to readers. If the writer is too focused on making sure to show off the narrators lack of consideration or ethics, readers are more likely to be repelled and stop reading. There are two common solutions: First, allow the narrator to be clearly unpleasant, but make them (or make the plot) fascinating enough in some other way that the reader will stick with the story in spite of the lack of sympathy for the narrator; or second, get so thoroughly into the obnoxious/unpleasant/amoral character’s head that their actions appear perfectly reasonable and justified unless the reader stops to think about it (after all, few, if any, people see their own actions as unpleasant, obnoxious, or evil).
Neither approach is easy. The more clearly repulsive the narrator, the more difficult it is to provide a combination of plot and personality that is strong enough to overcome the reader’s dislike of the character. On the other hand, the more reasonable-seeming and natural the narrator’s rationalizations for their actions, the more likely it is that some readers will take them at face value, or even assume the writer intends the narrator’s behavior to be realistically justified but hasn’t quite pulled it off.
Thank you very much for sharing your insights into writing processes.