Negative Analogy

A negative analogy identifies the specific ways two compared things differ. While comparisons help us explain complex ideas, every analogy eventually breaks down because the two things are not identical. Recognizing these 'break points' ensures you do not mistakenly apply rules from one context to another where they do not belong. For example, electricity flows like water, but unlike water, a cut wire does not spill electricity onto the floor. Ask this: When someone uses a metaphor, what is one specific way the two things are different?

Most people focus on 'Positive Analogies'—the features that two things share. However, the philosopher Mary Hesse argued that understanding a model requires identifying the 'Negative Analogy': the features known to exist in the source but not in the target. If you ignore these, you risk over-extending the metaphor and making bad predictions.

Think about the common business advice: 'Treat your company like a family.' The positive analogy is the sense of belonging and care. The negative analogy is that families are unconditional, while employment is conditional on performance. Failing to see this negative analogy leads to toxic guilt when firing is necessary or entitlement when performance lags.

Another example is in urban planning. Planners might view a city as a 'machine' for efficiency. While both have systems and flows (positive), a machine is static and designed by one author, whereas a city is organic and evolves through millions of independent choices (negative). Treating a city purely as a machine leads to rigid, lifeless neighborhoods.

To use this, whenever you hear a comparison, explicitly list three ways the source and target differ. This 'stress test' validates the boundaries of your mental model.

The concept of Negative Analogy originates from Mary Hesse's work on the role of models in science. She categorized analogical relationships into three types: positive (shared properties), negative (non-shared properties), and neutral (properties not yet known to be shared or distinct). Hesse argued that scientific discovery often occurs in the 'neutral' zone, but the validity of a model relies on clearly demarcating the negative analogy.

• Hesse (1966): A scientific model is only useful if it includes a 'negative analogy' that specifies where the model ceases to predict reality accurately. [1]
• Bartha (2019): Evaluating an analogy requires weighing the relevance of the positive analogy against the critical severity of the negative analogy. [2]
• Gentner (1983): While structural mapping emphasizes the alignment of relationships (positive analogy), high-level structural mismatches (negative analogy) are critical for distinguishing literal similarity from metaphor. [3]

Focusing too heavily on negative analogies can kill creativity or understanding in the early stages of learning. No analogy is perfect; if you nitpick every difference immediately, you may miss the useful structural similarities that the metaphor was intended to highlight.