Issue position argument mapping

Turn a tangled multi-question deliberation or transcript into an inspectable typed map by chaining shipped moves, surfacing the open issues, attaching rival positions to each, and running a pro/con pass on each position, so a sprawling debate becomes a navigable structure.

Flow

graph LR
  nthinkissuetree["Issue Tree"]
  nthinkdecisionoptionreview["Decision Option Review"]
  nthinkargumentmapping["Argument Mapping"]
  nthinkissuetree --> nthinkdecisionoptionreview --> nthinkargumentmapping

Each step passes a compressed artifact to the next. The numbered list below is the same chain in text.

The chain

1. think-issue-tree
2. think-decision-option-review
3. think-argument-mapping

Job: turn a tangled multi-question deliberation or transcript into an inspectable typed map - the open issues, the rival positions answering each, and the arguments for and against each position - using shipped moves rather than a standalone notation.

Use when: a discussion (a design debate, a strategy meeting, a long comment thread) has sprawled across several entangled questions and you want to see its structure: what is actually being asked, what answers are on the table, and what supports or attacks each. Especially useful on a transcript after the fact.

Why a recipe, not a skill: IBIS (Issue-Based Information System; Kunz and Rittel, 1970) has exactly three node types, and each is a move the library already ships. Issues (the open questions) are think-issue-tree’s decomposition with the MECE constraint relaxed. Positions (rival answers to a question) are the option-enumeration front of think-decision-option-review - the same place QOC, IBIS’s design-rationale sibling, already folds. Arguments (the pro/con layer) are a strict subset of think-argument-mapping (tier S), which additionally excavates co-premises and flags weak links that the flat IBIS argument layer does not. The only residue no shipped skill owns is the link grammar itself - positions may only answer issues, arguments may only attach to positions - and that is the wiring diagram of a chain, not a separable cognitive move. A sequence of existing skills produces the IBIS artifact, which is the definition of a recipe here, on the pattern of kepner-tregoe and pdca-a3. (It is not a fold, because no single shipped skill’s mode covers all three node types - argument-mapping has no issues or positions, issue-tree has no positions or arguments - so any single foldInto would misrepresent the method. The standalone-notation case is also where the instrument literature, Isenmann and Reuter 1997 and Shipman and Marshall 1999, documents IBIS tools failing; the positive evidence is all facilitated group practice, the facilitation wall this library has never shipped through.)

Chain

1. think-issue-tree (skills/think-issue-tree/SKILL.md) - Issues
   • Surface the open questions the deliberation is contesting. Relax MECE: IBIS issues need not partition the space, they just need to be the real questions in play. On a raw transcript, pull them out by question rather than by clean decomposition.
   • Carry forward: the set of open issues, not a full partition tree.
2. think-decision-option-review (skills/think-decision-option-review/SKILL.md) - Positions
   • For each issue, enumerate the rival candidate answers. Use the option-enumeration front of the skill; you are listing positions, not yet scoring to a single winner (though you may, if the deliberation has reached a decision).
   • Carry forward: the positions answering each issue.
3. think-argument-mapping (skills/think-argument-mapping/SKILL.md) - Arguments
   • For each position, map the pro and con arguments and their inference structure. Argument-mapping does strictly more than the flat IBIS pro/con layer - it surfaces co-premises and flags weak inferences - so take that as added rigor.
   • Carry forward: the arguments for and against each position.
   • Recurse: where an argument itself raises a new open question, add it as an issue and loop.

Composite artifact

A typed deliberation map: the issues, the positions answering each, and the arguments supporting or attacking each position - assembled from shipped moves, navigable where the raw debate was tangled. The map makes the structure of a sprawling discussion inspectable: which questions are still open, which answers have support, where the disagreement actually sits.

When NOT to use

When the deliberation is really one argument whose inference structure you want to test, use think-argument-mapping alone - you do not need the issue and position layers. When you are deciding among options on a single question, use think-decision-option-review alone. The recipe earns its weight only when there are several entangled questions to untangle at once. And it maps a deliberation; it does not adjudicate it - reaching a verdict on any one issue is the decision skill’s job, run on that issue.

Token discipline

Carry only the compressed artifact between steps - the issue list into the position step, the positions into the argument step - never the full working notes of the prior skill. On a long transcript, surface the top few issues first and map those; do not attempt an exhaustive node-by-node capture, which is the documented failure mode of IBIS tools (the capture overhead exceeds the value).