Voice and POV

POV options

• First person.Intimate, biased, audible. The narrator's voice is the prose's voice. Excellent for unreliable narrators or for character-driven books where interiority matters more than scope.
• Third person limited.The default of modern commercial fiction. We see only what one POV character sees per scene; we hear their thoughts. Allows voice flexibility (the prose can be more polished than the character's actual interior monologue).
• Third person omniscient. A narrator who knows everything dips in and out of multiple heads. Powerful for sweeping ensemble work; risky in a modern-realist register because readers are trained to expect one POV at a time.
• Close third.Third limited so close it reads almost like first. The narrator never breaks frame or uses words the POV character wouldn't use.
• Second person.“You walk into the room.” Hard to sustain at novel length. Useful in short stories, in specific stylized effects, and in instructional or confessional prose.
• Multi-POV. Multiple POV characters rotating across scenes; common in long-running epic fantasy and sprawling literary novels. Each POV section should feel distinctly voiced.

Voice levers

Voice is built from small, consistent choices. The four most load-bearing:

• Diction. Latinate vs. Anglo-Saxon. Formal vs. conversational. Specific vs. abstract. A biblical-cadence prose stylist might lean Latinate, abstract, and rolling. A pared-down minimalist might lean Anglo-Saxon, concrete, and declarative. Both extremes are voice. So is everything in between.
• Rhythm. Sentence-length variance. Short sentences hit. Long sentences flow. A page of all-medium sentences reads as competent prose with no voice.
• Cadence.Where you place the stress in a sentence. Late-stress sentences (“He learned, after many years, that the answer was simple”) sound literary. Early-stress sentences (“The answer was simple”) sound modern.
• Negative space. What you leave out. Voice often lives in the choice not to explain, letting the reader infer the emotional weight.

Common pitfalls

• Head-hopping. Slipping between POVs mid-scene without a structural break. Reads as careless; confuses the reader.
• Voice flattening across multi-POV. Three characters whose internal monologues all sound like the author. Each POV needs its own diction and rhythm.
• The author's voice leaking into a character's.A child character who suddenly uses a word the author would use but the character wouldn't.

Pendraic notes

The Anchors layer captures voice as a structured contract: register, diction sliders, rhythm parameters, taboos, reference authors. Pendraic enforces this on every AI generation via post-generation conformance scans, so a rewrite of a paragraph stays within your voice envelope instead of drifting toward the model's default register.