Person-less prose is writing from which the person has been removed. Not through violence but through the efficiency of processes that produce competent text faster than any human can produce voiced text. The term was introduced by composition scholar Mark Marino in 2025 research on students who had been exposed to ChatGPT. Marino found that students had 'developed some skill at imitating the cadence and rhythm of ChatGPT,' producing work characterized by 'complete sentences of similar length and structure' — technically correct, structurally sound, rhetorically adequate, and devoid of the irreducible presence Peter Elbow called voice. The phenomenon is not unique to AI; traditional academic writing often exhibits the same voicelessness, produced by students trained to suppress their own thinking in favor of formulaic structures. But AI accelerates and intensifies the pattern, making voiceless competence available at scale and eliminating the institutional friction that once made voice development unavoidable.
Person-less prose is the precise negation of everything Elbow championed. It is writing that fits the mold, meets the requirement, satisfies the rubric — and contains no trace of the consciousness that produced it. The sentences are grammatically correct. The structure is logical. The arguments cite appropriate sources. But no felt sense guided its production. No first-order discovery shaped its formulations. No surprise interrupted its smooth unfolding. The prose performs the appearance of thought without thought having occurred, and the performance is so convincing that external evaluators — teachers, managers, clients — often cannot distinguish it from genuine thinking.
The phenomenon predates AI but is intensified by it. Traditional five-paragraph essays, corporate memos drained of personality, academic papers written in the impersonal passive voice — all exhibit person-lessness produced by pedagogical and professional norms that reward conformity over discovery. What AI changes is the speed and scale. Before AI, producing person-less prose required the student to suppress her own voice manually, which was uncomfortable and left traces of resistance. After AI, person-less prose is generated by the machine, and the student's role is merely to review it. The suppression is complete because there was never a voice to suppress — the machine never had one.
John Warner's response to Elbow's death positioned AI as validation of Elbow's original critique: the formulaic 'school artifacts' traditional pedagogy demanded were always person-less, always mechanical simulations designed to satisfy an evaluative framework rather than to embody genuine thought. If a machine can produce them, they were never measuring what education claimed to measure. The validation is real, but it carries a darker implication: if the artifacts are now abundant and cheap, and if institutions continue to reward them, the incentive to develop voice collapses. The student who can generate an A-grade essay in thirty seconds has no external reason to spend three hours struggling to discover what she actually thinks.
The antidote is not the prohibition of AI tools but the redesign of what institutions measure. The teacher who grades questions rather than essays measures first-order capacity — the ability to identify gaps in understanding, to articulate what one does not know, to open spaces that require genuine engagement with material. Questions resist outsourcing because they depend on the specific configuration of a particular student's prior knowledge and confusions. The law school that evaluates through unassisted oral argument measures reasoning under pressure. The medical school that assesses through direct patient interaction measures embodied diagnostic capacity. These assessments cannot be gamed by person-less prose, because they require the presence of a person — thinking, judging, responding in real time to material that resists.
Mark Marino's research emerged from his position as Director of the Humanities and Critical Code Studies Lab at the University of Southern California. His 2024–2025 studies of students who had integrated ChatGPT into their writing workflows documented the unconscious imitation of machine cadence and the progressive thinning of individual voice. The term 'person-less prose' was coined to distinguish this phenomenon from plagiarism — the student had not copied the machine's words but had absorbed its rhythms, internalized its tonal register, and begun producing text that replicated the machine's voicelessness without deliberate effort.
Voicelessness is contagious. Extended exposure to AI-generated text causes unconscious imitation of its cadence and structure — writers begin producing voiceless text without realizing the shift has occurred.
Competence without presence. Person-less prose can be technically flawless, structurally elegant, and completely devoid of the signal that distinguishes one writer from another — the gap between quality and voice becomes visible.
AI reveals prior person-lessness. Traditional academic writing was often already voiceless, produced by students trained to suppress their thinking — machines simply automate what institutions were already demanding.
External rewards drive voicelessness. When smooth artifacts receive high grades and rough discoveries receive low grades, rational students optimize for smoothness — assessment must change to reward presence over polish.
The antidote is voice development. Freewriting, the garbage draft, and the silence between prompts protect the first-order space where voice forms.