PASS Agentless passives

Xiao B28 MFTE PASS child of PASS_ALL

Definition

Passive voice constructions with BE auxiliary where no agent is expressed (e.g., “the ring was found”, “he was taken prisoner”).

Parent

PASS_ALL

Detection Rules

mfte

MFTE tags PASS on the VBN (past participle) following a form of BE. Anchor on last (the participle) to match MFTE’s tagging convention and enable tag refinement (refines: VBN).

cql[lemma="be" & pos="VB.*"] [pos="VBN"]

Requires: lemma, pos

Anchor: last

Refines: VBN

Verb with auxpass dependent. Agentless if no “by” + NP follows. Requires post-filter to exclude cases with “by” agent phrase.

semgrex{pos:/VB.*/}=verb >auxpass {lemma:be}

Requires: lemma, pos, dep

pybiber

Auxpass dependency label. pybiber finds all auxpass tokens then excludes those with “by” 2-3 tokens ahead (which become PASSBY).

cql[dep="auxpass"]
combine: _ & !PASSBY

Requires: pos, dep

BE auxiliary + past participle. Less accurate — false positives from progressive BE + adjective. Requires post-filter to exclude by-passives.

cql[lemma="be"] [pos="VBN"]

Requires: lemma, pos

Normalization

Per finite_verbs

Examples

The ring was found in the river.

He was taken prisoner and brought before the king.

The letter was written in haste.

He must have been burgled.

Source: le_foll_2024

They need to be informed.

Source: le_foll_2024

He was found out.

Source: le_foll_2024

When were they arrested?

Source: le_foll_2024

Non-examples

The door was open.

Sources

  • biber_1988 — Biber, Douglas (1988) : Variation across Speech and Writing
  • mfte — Le Foll, Elen & Shakir, Muhammad (2023/2025) : Multi-Feature Tagger of English (MFTE) — Python version
  • pybiber — Brown, David West & Reinhart, Alex (2026) : pybiber — Python package for linguistic feature extraction and Multi-Dimensional Analysis
  • xiao_2009 — Xiao, Richard (2009) : Multidimensional analysis and the study of world Englishes
  • bohmann_2019 — Bohmann, Axel (2019) : Variation in English Worldwide: Varieties and Genres in a Quantitative Perspective

Notes

D5 loading .43. MFTE merges with by-passives (PASSBY). Agentless passives suppress the agent, making prose more abstract and impersonal.