Dimensional Modification for Hypothetical Case Generation

14 concepts9 papers1 proposal

The Limits of Static Rules

When evaluating a complex argument or statutory rule, traditional logic often stumbles against "open-textured" concepts—terms that are intentionally or inherently ambiguous. In these domains, applying static rules to a fixed set of facts is insufficient. To truly understand a rule, one must understand where it ceases to apply.

As highlighted in early foundational discussions on AI and law (such as AI and Legal Reasoning and Al and), legal reasoning relies heavily on the strategic use of hypothetical scenarios ("hypos") to probe these conceptual boundaries. Historical precedents are often too limited in scope to test every edge case or expose latent weaknesses in an interpretation. Without a way to generate synthetic variations of a case, reasoning systems cannot anticipate "slippery slope" effects or identify unintended liabilities caused by vague statutory definitions—a problem illustrated by Brown in A Critique of the Proposed West Virginia Criminal Code.

To stress-test an argument, practitioners need a systematic way to construct hypothetical scenarios that push propositions to their logical limits.

Representing Cases via Dimensional Frames

How can a computational system generate meaningful hypotheticals without simply inventing random scenarios? The solution lies in Constrained Example Generation (CEG). As proposed by Rissland in Legal Hypotheticals, effective hypotheticals are rarely built from scratch; instead, they are generated by applying a sequence of strategic modifications to an existing "seed" case.

To automate this process, scenarios are represented as structured frames containing modifiable dimensions. A dimension represents a specific factual variable that directly influences the strength of a position. For example, in trade secret law, dimensions might include the number of employees who know the secret, or the security measures taken to protect it.

By formalizing cases as sets of modifiable dimensions, Rissland et al. demonstrated in Explaining and Arguing With Examples that a system can perform "case deformation." By incrementally shifting values along these dimensional axes, the system changes the factual balance of the case, allowing it to systematically map out exactly where a rule or interpretation breaks down.

Heuristics for Case Deformation

Case deformation is guided by specific rhetorical and pedagogical goals, rather than random perturbation (as explored in Examples in Legal Reasoning). In the HYPO system, Rissland et al. formalized five key generation heuristics in Hypotheticals as Heuristic Device to transform a seed case into a structured "Heuristic Constellation" of hypotheticals:

  1. Adjusting specific factual values: Incrementally strengthening or weakening a single factor to test argument sensitivity.
  2. Extreme modifications: Pushing a dimension to its absolute logical limit to reveal controversial or absurd consequences of a legal interpretation.
  3. Exploring conflicting dimensions: Moving multiple variables in opposite directions simultaneously to evaluate competing principles.
  4. Manipulating cases toward a desired outcome: Intentionally altering facts to flip the expected conclusion, helping an arguer anticipate opposing viewpoints.
  5. Creating "near misses": Generating scenarios where most, but not all, prerequisites for a specific dimension are satisfied.

As detailed by Ashley et al. in Compare and Contrast: A Test of Expertise, isolating these precise factual gaps allows the system to explore how minor real-world changes alter the viability of a position.

Why It Works: Mapping the Conceptual Space

Dimensional modification works because it transforms the task of evaluating a discrete, static rule into the task of navigating a continuous conceptual space.

Traditional keyword searches and strict rule-based systems fail when encountering novel situations because they cannot evaluate how sensitive an argument is to minor factual changes. By generating a cluster of hypothetical variations around a real-world situation (see also Defining), dimensional modification allows a reasoning system to probe the boundaries of a concept.

Instead of treating a case as an isolated data point, this approach treats the case as a starting position on a multi-dimensional map. Moving along the axes of this map reveals the transition states between a winning and a losing argument, uncovering hidden assumptions and exposing the long-term consequences of a given interpretation before a final decision is ever reached.

Key Foundational Work

Examples in Legal Reasoning: Legal Hypotheticals (1983) — contributed 2 concept elements to this direction.
Examples in Legal Reasoning: Legal Hypotheticals (1983) — contributed 2 concept elements to this direction.
A Critique of the Proposed West Virginia Criminal Code (1977) — contributed 1 concept elements to this direction.
History & citationVersion 1 · 2026-10-04

Current · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_hypothetical_scenario_generation_f09a1eac,
  title = {hypothetical scenario generation},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/hypothetical-scenario-generation/rdr_f09a1eac2f634121806b9d6ea1cc74a0},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-22},
}

This exact version (version 1, frozen text):

@misc{lacuna_hypothetical_scenario_generation_f09a1eac_v1,
  title = {hypothetical scenario generation},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/hypothetical-scenario-generation/rdr_f09a1eac2f634121806b9d6ea1cc74a0/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-22},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake