Is Luke the author of a Gospel and the Acts of the Apostles?
Open accessThe DOI matches its Crossref record; 38 of the 61 references it cites were found in a registry or an index, and 13 are not checked yet; none of the 21 sources its claims cite is checked for support yet.
Abstract
According to Christian tradition, Luke is credited with authoring a Gospel and the Acts of the Apostles, even if his name does not appear in either book, both originally written in Koine Greek. Several biblical scholars assume that both texts were written by a common author, while others deduce the presence of two authors. Different studies have been found to support either finding, some based on qualitative evaluation, while a few others consider the occurrence frequency differences between the two books. To propose an enhanced quantitative analysis, this study is grounded on two recent authorship attribution models. The Burrows’ Delta, applied with eleven different feature sizes, demonstrates common authorship. An author verification model confirms this finding. The following experiments consider several stylistic representations, feature sizes, and distance functions to confirm that Luke is the true author of both books.
Checked 2026-10-05 · 13 references not checked yet
Overview
CoChat Summary
## TL;DR
The paper applies stylometric authorship attribution and verification methods to Luke and Acts in Koine Greek and argues that they have a common author. Across its tests, Burrows’ Delta and a verification model both consistently support Luke as the author of both books.
## Key findings
- Burrows’ Delta was run with 11 feature-set sizes and “always indicates common authorship” for Luke’s Gospel and Acts.
- In the verification experiments, the model was run with 8 stylistic representations, 6 similarity/distance measures, 10 feature sizes, and 1,000 iterations per feature size; it “always confirms” common authorship.
- The corpus used for the main comparison included Luke (19,482 tokens) and Acts (18,450 tokens); the paper also reports that the two books together are about 28% of the New Testament.
- The paper reports that reliable assessment generally required 100 or more features; with fewer than 200 features in the verification table, results were said to need caution.
## Methods
- Design: Quantitative authorship attribution/verification study comparing Luke and Acts against other New Testament books and impostor texts.
- Data/sample: Koine Greek texts downloaded from Bibelwissenschaft.de; the main corpus included the Gospels of Matthew, Mark, Luke, John, Acts, four Pauline letters, and Revelation, with chapter/token counts listed in Table 1. The verification model chunked Luke and Acts into 5,000-token segments.
- Main analysis: The study used Burrows’ Delta with multiple feature sizes, then an authorship verification model (Koppel & Seidman, 2018) with multiple stylistic representations (MFW, n-grams, truncation words), several distance/similarity measures (including cosine, MinMax, Dice, Manhattan, Tanimoto, Euclidean), and random selection of 50% of features over 1,000 iterations.
## Limitations
- The paper explicitly says this study “must be viewed as a first step” and that applying modern statistical modeling to Christian literature is “a largely open domain.”
- The author notes that the number of texts used is limited, so the probability estimates “must [be taken] with a grain of salt.”
- The text makes evident that the study depends on specific corpus choices and preprocessing (e.g., chapter-based Greek texts, verse-number/title removal, and a limited set of impostor texts), which may constrain generalization.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 Valid · 2 Not valid · 14 Not checked · 48 Own results and interpretations
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“For example, the word Παῦλος (Paul, noun, masc. sing. nom.) appears 55 times in Luke's writings, but only 9 occurrences appear with the other writers.”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“In other circumstances, such as the last portion of Acts (Paul travelling at sea), the system found some similarity with the beginning of John's Gospel.”
hedgedstates an associationSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“For example, when applying the cosine similarity with 50, 75, 100, and 150 features, the result vector has a first component that is lower than 0.8 (the d threshold).”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
In the paper
“Third, a topical relationship clearly appears between the contents of both books in their justification of preaching not to the Jews, but to the Gentiles”
What the sources say
- Not checked — no full text of the sourcecites Burkett 2019 source
boostedstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Applying this variant, the Z score associated to Παῦλος (Paul) is now 7.37.”
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
22 more claims
Create a free account to see all 64 claims and every source verdict.
- Probably not a substantial claim. 20
- It may state two separate claims as one. 21
- The quote only partly states this claim. 2
Citations
Cites
Introduction to information retrieval
(2009) · Choice Reviews Online · doi:10.5860/choice.46-2715 · Cited by 12,278
Introduction to Information Retrieval
Manning, C. D., Raghavan, P., & Schütze, H. (2008) · doi:10.1017/cbo9780511809071 · Cited by 11,288
Machine learning in automated text categorization
Sebastiani, F. (2002) · ACM Computing Surveys · doi:10.1145/505282.505283 · Cited by 7,926
Analyzing Linguistic Data: A Practical Introduction to Statistics using R
Baayen, R. H. (2008) · doi:10.1017/cbo9780511801686 · Cited by 3,253
An Introduction to Statistical Learning: with Applications in R
James, G., Witten, D., Hastie, T., & Tibshirani, R. (2021) · Springer Texts in Statistics · doi:10.1007/978-1-0716-1418-1 · Cited by 1,755
Delta: A Measure of Stylistic Difference and a Guide to Likely Authorship
Burrows, J. (2002) · Literary and Linguistic Computing · doi:10.1093/llc/17.3.267 · Cited by 627
Zheng, R., Li, J., Chen, H., & Huang, Z. (2006) · Journal of the American Society for Information Science and Technology · doi:10.1002/asi.20316 · Cited by 519
Shakespeare, Computers, and the Mystery of Authorship
(2009) · Cambridge University Press eBooks · doi:10.1017/cbo9780511605437 · Cited by 284
Attributing Authorship: An Introduction
Love, H. (2002) · doi:10.1017/cbo9780511483165 · Cited by 235
Determining if two documents are written by the same author
Koppel, M., & Winter, Y. (2014) · Journal of the Association for Information Science and Technology · doi:10.1002/asi.22954 · Cited by 227
Metadata
- Authors
- Savoy, J.
- Affiliations
- University of Neuchâtel (CH)
- Corresponding author
- Savoy, Jacques
- ORCID
- Savoy, Jacques
- Venue
- Digital Scholarship in the Humanities
- Publisher
- Oxford University Press (OUP)
- Published
- 2026-09-21
- ISSN
- 2055-7671 (ISSN-L)
- 2055-768X
- OpenAlex
- W7213529827
- Cited by
- 0
- References
- 56
- FWCI
- 0.00
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- Digital Scholarship in the Humanities (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- arXiv (Cornell University) (cc-by)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Authorship Attribution and Profiling
- Social Sciences › Arts and Humanities › Literature and Literary Theory › Digital Humanities and Scholarship
- Social Sciences › Arts and Humanities › Religious studies › Biblical Studies and Interpretation
- Keywords
- Gospel of Luke, Acts of the Apostles, authorship attribution, Burrows' Delta, author verification, stylistic features, Koine Greek, New Testament, quantitative stylistics
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-01, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-01 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Is Luke the author of a Gospel and the Acts of the Apostles? https://app.cochat.ai/papers/10.1093/llc/fqag118
- The paper
- Savoy, J. (2026). Is Luke the author of a Gospel and the Acts of the Apostles? Digital Scholarship in the Humanities. https://doi.org/10.1093/llc/fqag118
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.