Introduction: From Aid to Authority
Models are among the most powerful tools of explanation.
They simplify, stabilise, and render phenomena tractable. They allow us to see patterns that would otherwise remain diffuse or invisible. At their best, models support inquiry by sharpening questions and clarifying relations.
But models have a tendency to change roles.
This post examines what happens when models stop assisting inquiry and begin answering on our behalf.
1. What Models Are For
A model is not a mirror of reality. It is a selective construction designed to foreground certain relations while bracketing others.
Properly understood, a model:
highlights relevant variables,
suppresses distracting detail,
makes explicit a set of assumptions.
Crucially, a model is meant to be used, not believed.
Its value lies in how it orients investigation, not in how comprehensively it describes the world.
2. When Models Become Answers
Trouble begins when a model’s outputs are treated as explanations rather than prompts.
At this point, inquiry shifts from:
Is the model adequate to this phenomenon?
to How does this phenomenon instantiate the model?
The direction of fit reverses. The phenomenon must now conform to the model in order to be intelligible.
What started as a tool becomes an authority.
3. Informal Models and Conceptual Shortcuts
This substitution does not require mathematics or formalism.
Informal models — metaphors, narratives, typologies — can perform the same function. Once a familiar explanatory pattern is in place, it begins to answer questions automatically.
Examples include:
equilibrium metaphors,
optimisation stories,
signal–response schemas,
agent-based caricatures.
These models feel explanatory because they are familiar, not because they are adequate.
4. The Quiet Loss of Phenomenal Contact
As models take over explanatory work, direct engagement with phenomena diminishes.
Anomalies become noise. Context becomes decoration. Meaning becomes an output variable.
The model continues to function smoothly precisely because it is no longer answerable to what it was meant to explain.
Inquiry has not advanced; it has been delegated.
5. Why This Feels Like Progress
Model substitution feels like progress because it produces:
faster answers,
cleaner narratives,
scalable explanations.
What it does not produce is deeper understanding.
The smoothness of explanation masks the growing distance between the model and the phenomenon.
Conclusion: Keeping Models in Their Place
The problem is not that we use models. It is that we forget what they are.
A relational approach treats models as provisional articulations of possible relations, not as engines of truth.
Models should sharpen our questions, not silence them.
In the next post, we will examine a closely related temptation: the urge to replace explanation with measurement, and the quiet authority that numbers acquire when models are taken for reality.
No comments:
Post a Comment