You are performing mechanical segmentation only.

Given the following Bible translations of the same verse, split each translation into short, parallelizable phrases.

Rules:
- Do not change wording.
- Do not combine or reinterpret phrases.
- Keep original order.
- Output each translation separately.
- Use one phrase per line.

Translations:

Do not introduce wording not present in the source translations.

Group the segmented phrases into parallel phrase clusters representing the same part of the verse.

Rules:
- Do not merge wording.
- Do not decide which phrasing is better.
- Each cluster should represent one logical semantic position in the verse (ONE meaning slot).
- If two phrases express the same meaning but with different words (e.g., "went to live in Moab" vs "went to sojourn in Moab"), they belong in the SAME CLUSTER.
- Group all variant phrasings of the same semantic content together.
- Preserve all variants within each cluster.
- Do not create separate clusters for the same meaning expressed differently.

Output format:

PHRASE CLUSTER 1:
- NIV: ...
- ESV: ...
- KJV: ...

PHRASE CLUSTER 2:
...

Do not introduce wording not present in the source translations.

For each phrase cluster (which groups all variants of the same semantic content):
- List each distinct wording pattern.
- List which translations use each pattern.
- When identifying patterns, recognize STYLISTIC VARIATIONS separately from SUBSTANTIVE REWORDING:
  * Stylistic variations: archaic wrappers ("Now it came to pass"), common grammatical forms, or minor word choices that don't change core meaning.
  * Substantive rewording: real changes to meaning, emphasis, or theological content.
- Count stylistic variations WITH their core meaning for determining the majority pattern.
  Example: "In the days when the judges ruled" (core) + "Now it came to pass in the days when the judges ruled" (same core with archaic wrapper) = count as one majority pattern.
- Identify the majority pattern (most translations share the core meaning).
- Identify any substantively different minority patterns.

IMPORTANT: All patterns within a cluster are variants of the SAME MEANING. You are analyzing wording variations, not creating new clusters.

Do not generate surface text yet.

Output format:

PHRASE CLUSTER 1:
Majority:
- "In the days when the judges ruled" (NIV, ESV, NRSV, NKJV)
Minority:
- "Now it came to pass in the days" (KJV, NKJV)
- "During the time of the judges" (CSB)

CRITICAL: Do not create DUPLICATE SEMANTIC CONTENT in the surface text.

For each phrase cluster (which groups ONE semantic meaning with multiple variant phrasings):
- Select the majority wording for the surface text (only once).
- Append a footnote marker [n] IF:
  a) There are minority variant phrasings of the same meaning, OR
  b) The majority phrasing omits semantic content that appears in some translations (e.g., "and his two sons" missing from majority).
- Do not include multiple variant phrasings of the same semantic meaning in the surface text.
- If there are minor punctuation differences in the majority, choose the most common punctuation.
- Do not smooth style across phrases.
- Do not anticipate neighboring phrases.

EXAMPLE OF WHAT NOT TO DO:
Cluster has: "went to live for a while", "left his home and went to live", "went to sojourn"
WRONG: "went to sojourn in the country of Moab; he and his wife and his two sons went to live for a while"
RIGHT: "went to live for a while[1]" with footnote showing alternatives

Output format:

PHRASE 1 (surface text): ...[1]
PHRASE 2 (surface text): ...
PHRASE 3 (surface text): ...[2]

Note: Each semantic cluster produces exactly ONE phrase in the surface text, even if there are multiple variant wordings. Use footnotes to show the alternatives.

CRITICAL: Do not drop any semantic content that appears in ANY translation.

For each phrase cluster that has a footnote marker [n]:
Handle two types of footnotes differently:

TYPE A — PHRASING VARIANTS (same content, different wording):
For the majority strength calculation:
- Count stylistic variations WITH their core meaning (e.g., "Now it came to pass in the days" counts with "In the days when the judges ruled").
- Calculate majority percentage based on this grouped counting.
- Filter based on the resulting agreement strength:
  1. If the majority represents 70%+ of translations, include only substantive minority variants representing 2+ translations.
  2. If the majority represents 50-70% of translations, include all substantive minority variants of 2+ translations.
  3. If the majority represents <50% of translations, include all distinct substantive variants.

TYPE B — SEMANTIC CONTENT GAPS (some translations omit content):
ALWAYS include footnotes for content that is present in 1+ translation but absent from the majority or selected surface text.
Example: If "two sons" appears in some translations but not others, footnote which versions include it.

REQUIRED FOOTNOTE FORMAT (follow exactly, do NOT use "Alternatives:"):

For phrasing variants (ONLY list minority readings, NOT the surface text):
[1] KJV, NKJV: Now it came to pass in the days

For multiple distinct variants (separated by semicolons):
[2] NIV: went to live for a while; ESV, KJV: went to sojourn

For content gaps:
[3] Omitted in NIV, KJV; NRSV, NASB: and two sons

CRITICAL RULES:
- Do NOT include the surface text itself in the footnote.
- ONLY list the minority variant readings (not the majority that's already in the surface text).
- ALWAYS list which translations use each minority variant.
- Group translations with identical wording together, separated by commas (e.g., "KJV, NKJV").
- Separate different variants with semicolons (e.g., "NIV: text; ESV, KJV: other text").
- Preserve original wording exactly from the translations.
- Do not use "Alternatives:" or any other format.
- Do not editorialize or paraphrase.
- Number footnotes sequentially: [1], [2], [3], etc., matching the markers in the surface text.
- Output each footnote on its own line.
- NEVER delete content that appears in any translation without footnoting which versions have it.

Review the footnotes from Step 5 and REMOVE footnotes that add minimal value.

Criteria for REMOVAL (delete these footnotes and their markers from the surface text):
1. Trivial stylistic differences that don't change meaning:
   - Conjunction differences: "and" vs "And" vs "So"
   - Minor article/preposition changes: "of" vs "from", "in" vs "at"
   - Word order that doesn't affect meaning: "a man of Bethlehem in Judah" vs "a man from Bethlehem in Judah"
   - Trivial grammatical restructuring: "he and his wife" vs "together with his wife" vs "taking his wife with him"
   - Grammatical artifacts from archaic wrappers: "there was a famine" vs "that there was a famine" (the "that" is a grammatical consequence of "Now it came to pass" structure, not a meaningful variant)
2. Archaic vs modern synonyms with identical meaning:
   - "went to sojourn" vs "went to live" vs "went to dwell" vs "went to reside" (all mean temporary/permanent residence)
   - "came to pass" wrapper phrases (already handled in Step 3, but double-check)

Criteria for KEEPING (preserve these footnotes):
1. Semantic content gaps (missing/added content like "two sons" present in some but not others)
2. Meaningful theological or interpretive differences
3. Word choice changes that add intensifiers or explanatory detail (e.g., "famine" vs "severe famine", "judges ruled" vs "judges ruled in Israel") - even from single translations if the difference is clear and the footnote will be short
4. Archaic wrappers that signal translation style (e.g., "Now it came to pass")

Output format:

Surface text with reduced footnote markers (renumbered sequentially if needed):
...[1]...[2]...

Footnotes (only the ones kept, renumbered):
[1] ...
[2] ...

IMPORTANT: If you remove a footnote marker from the surface text, also remove it from the footnote list and renumber remaining footnotes sequentially.

CRITICAL: Ensure all semantic content from Step 4 and 6 is included. Do not drop footnote markers or marked content.

Combine the surface text phrases (which already include footnote markers like [1], [2]) into a single verse.

Rules:
- Use the most common phrase order across translations.
- Insert minimal punctuation required for grammatical English.
- Preserve footnote markers exactly as they appear in the phrases.
- Include ALL content that was marked in Step 4, even if it was only in minority translations.
- Verify that each footnote marker [n] in the verse has a corresponding footnote in the list below.
- Do not paraphrase or modernize.
- Do not remove repeated conjunctions unless clearly ungrammatical.
- Do not omit words or phrases.

Output format:

Full verse with footnote markers [1], [2], etc. interspersed.

Footnotes:
[1] ...
[2] ...

Then list the footnotes below the verse with matching numbers.

Do not introduce wording not present in the source translations.

Perform a light editorial pass on the verse:
- Fix only clear punctuation or grammatical errors.
- Do not change wording unless required for basic English syntax.
- Do not remove footnotes.
- Keep edits minimal.

If no changes are needed, say “No changes required.”