
Was There a Q? — the Synoptic Problem
A small static explainer site on the Synoptic Problem: did Matthew and Luke
independently use Mark and a lost sayings source, Q, or did one of them read
the other?
It sets out the problem and the two broad answers, the data (how much the
Gospels share, the order of their material, and the Greek texts side by side,
marked by which Gospels share each word), four main hypotheses and eight minor
ones with diagrams, the arguments grouped by question (each with its evidence,
the case for and against, the rejoinder and where it stands, plus a hypothesis
selector that shows how each argument bears on a chosen hypothesis), who holds
which position, claims to retire, experiments run for the page, an assessment that takes a position, and a bibliography.
It is a sibling of Angry or Compassionate? Mark 1:41
and follows its layout and conventions.
Files
index.html page shell: masthead, section bar with the hypothesis picker, empty sections
styles.css all styles (light and dark, phone layout, print)
app.js renders every section from content/ at load time; no framework, no build step
content/
meta.json title, lede, the two answers at the top, section intros, basis legend, credits
phenomena.json counts, terms, order tables (Markan and double tradition), takeaways
synopsis.json the Greek exhibits (generated: see below)
chapters.json verses per chapter, for the order charts (generated: see below)
models.json the four main hypotheses, othersIntro and the other proposals, with diagram data
arguments/ one file per group, loaded in this order: P, I, C, D, Q, M, S
positions.json who stands where, in period groups
history.json how the question developed (timeline)
corrections.json claims to retire
research.json experiments run for this page: each with question, method, results, optional figure and table(s), reading, limits, files; plus planned tests
assessment.json the site's own assessment (optional: until it exists the section says "Not yet written")
bibliography.json every work cited: key, citation, optional url, basis
tools/
build_synopsis.py builds content/synopsis.json and content/chapters.json
exhibits.json hand-written specification of each exhibit (passages, title, lede, points, usedBy)
data/ MorphGNT SBLGNT data for Matthew, Mark and Luke (unmodified; see tools/data/README)
kloppenborg_test.py experiment E1 (verbal agreement); kloppenborg_test.txt is its output
research_stylometry*.py experiment E2 (style of Q and special Matthew), two rounds, with outputs
research_overlaps.py experiment E3 (Luke's words in the Mark-Q overlap passages)
research_minor_agreements.py experiment E4 (a checkable list of minor agreements)
research_hide_mark.py experiment E5 (recovering Mark from Matthew and Luke)
research_order.py a test of who rearranged the double tradition; failed its check, not on the page
research_review_checks.py the reviewers' checks, with research_e1_verses.json (verse-by-verse judgements)
research_common.py loader shared by the scripts; they need the public/data folder of Synoptic Canvas
review/ the two adversarial review reports (AUDIT.md, REVIEW.md); they use the
experiments' first numbering (E1-E7), before the page was condensed
assets/icon.svg repository icon
Editing the content
Fix a date, add a reference or change wording in content/; the page is
rendered from these files at load time, so no layout code needs to change.
- Text fields accept plain text,
*italic*, Greek (tagged automatically for
the Greek font) and [[C3]], which becomes a link that opens argument C3
([[E1]] links to experiment E1 in the same way).
A link to an argument that does not exist yet is shown greyed out, not broken.
- Bibliography entries are linked automatically wherever a reference contains
an entry's
key (e.g. Goodacre 2002, ch. 8). The longest key wins, so
Goodacre 2002a never links to Goodacre 2002; keys with &, et al.,
diacritics or letter suffixes work as written. Each entry's anchor is
#bib- plus the key in lower case with accents removed, & as and and
other punctuation as hyphens (Müller & Omerzu 2018 → #bib-muller-and-omerzu-2018).
- Every item can carry a
basis (read, abstract, reported, unchecked),
shown as a small marker and explained in the legend at the top and bottom
of the page.
- A missing or invalid file does not break the page: its section shows
"Not yet written".
Links: #arg-C3 opens an argument, #exp-E1 jumps to an experiment, #ex-john-preaching opens an exhibit,
#bib-goodacre-2002 jumps to a bibliography entry, and ?model=FH traces a
hypothesis through the arguments.
Regenerating the Greek synopsis
content/synopsis.json and content/chapters.json are generated; do not edit
them by hand. To change an exhibit, edit tools/exhibits.json and run, from
the repository root:
python3 tools/build_synopsis.py # rewrite content/synopsis.json and content/chapters.json
python3 tools/build_synopsis.py --check # also print each exhibit as text with its agreement marks
python3 tools/build_synopsis.py --dump Matt 3:11-12 # numbered tokens of a passage, for writing blocks
The script needs only Python 3. It takes the Greek text and lemmas from the
MorphGNT files in tools/data/, aligns the parallel columns word by word, and
marks each word with the Gospels that share its exact form (g) and its
lemma (gl). How it aligns and counts is described at the top of the script
and summarised on the page under "How counted". An exhibit may set
"aligned": true or false to force or prevent verse-by-verse rows; by
default rows are aligned only when the columns have the same number of verses
and most rows share wording.
Running locally
No build step. Serve the folder with any static server, for example:
python3 -m http.server 8000
then open http://localhost:8000. (Opening index.html directly from disk
will not work, because the page fetches the files in content/.)
Credits and licences
- Greek text: The Greek New Testament: SBL Edition (SBLGNT), © 2010 Society
of Biblical Literature and Logos Bible Software, licensed
CC BY 4.0.
- Lemmas, normalised forms and parsing: MorphGNT SBLGNT
(James K. Tauber and contributors), licensed
CC BY-SA 3.0. The data
files in
tools/data/ are unmodified copies; content/synopsis.json, which
is derived from them, is shared under the same licence.
- English renderings of the Greek are working translations made for this page.
- Web font: Gentium Book Plus (SIL Open Font License), loaded from Google Fonts.