Absurdum explicandum

Following my previous past on different modes of explanation, I was interested in how an audience would decide that an explanation was inadequate, did not make sense or was full of errors. Avoiding the language of truth and falsity, I asked NotebookLM: “Please mirror these five categories of explanation with examples of explanations that most people would regard as absurd.”

The platform responded with examples of absurd explanations for the design decision of incorporating a large window in a living room, mirroring the five categories discussed in that post. Here’s its edited response.

Formal Logic

This approach previously used deductive reasoning from stated premises. An absurd logical explanation would involve faulty premises or illogical deductions:

• “1. Premise 1: All furniture dislikes sunlight intensely [this is a false and generally unsupported premise]. 2. Premise 2: Large windows allow maximum sunlight to enter a room. 3. Deduction: Therefore, to ensure furniture experiences maximum discomfort, large windows should be included in the living room design.”

This is absurd because the initial premise contradicts common understanding about the purpose of living spaces and the properties of sunlight. The logical deduction, while following from the premises, leads to a nonsensical design outcome based on this flawed foundation. This plays with a kind of inverted logic, similar to the paradoxes and contradictions sometimes explored in more abstract thought.

Rules-based system

This approach previously cited established design guidelines. An absurd explanation based on a “rule” would involve a nonsensical or irrelevant regulation:

• “Rule: All living rooms must be designed to facilitate the unobstructed passage of migratory birds through the internal space. Large windows are mandatory to allow birds to navigate freely.”

This is absurd because it invokes a completely fabricated and impractical “rule” that has no basis in architectural principles or human habitation. It introduces an element of the incongruous and illogical.

Precedents

This approach previously referenced successful historical examples. An absurd explanation based on precedent would cite an irrelevant or misinterpreted case:

• “Explanation: The inclusion of large windows in this living room is directly based on the successful design of medieval dungeons, which often featured small, high openings that, by extension, prove that any opening, regardless of size, is a time-honoured architectural tradition for all enclosed spaces. Therefore, exaggerating the size of these traditional openings is a logical continuation of this precedent.”

This is absurd because it draws a nonsensical parallel between the function and design principles of a living room and a dungeon. It misinterprets the significance of historical features and applies them out of context, creating an incongruous link. This also touches on the idea of misrepresentation discussed in the context of rationalism (in various source documents).

Metaphor

This approach previously used analogy to create understanding. An absurd metaphorical explanation would employ a bizarre or nonsensical comparison:

• “Explanation: The living room is a ‘giant internal nostril’, and the large windows are essential for it to ‘smell’ the outside world effectively. Just as large nostrils allow for a greater intake of scents, these expansive windows enable the room to fully experience the olfactory essence of the surrounding environment.”

This is absurd due to the nonsensical and illogical comparison between a living room and a nostril, and the attribution of a literal sense of smell to a building. The metaphor fails to create any meaningful connection and instead highlights the paradoxical idea of a room “smelling”.

Neural networks

This approach previously simulated how an LLM might explain the design based on statistical patterns. An absurd LLM “explanation” would highlight spurious correlations or nonsensical data associations:

• “LLM Explanation (simulated): Based on its training data, the model has identified a strong statistical correlation between the presence of ‘large windows’ in images tagged ‘living room’ and the simultaneous occurrence of ‘images of cats stuck in trees’ and forum discussions about ‘the existential dread of garden gnomes’. Therefore, the inclusion of large windows is likely intended to resonate with users who have an affinity for these statistically co-occurring concepts.”

This is absurd because it highlights nonsensical correlations in data that have no logical or practical relationship to the design of a living room. The “explanation” relies on the arbitrary co-occurrence of terms and concepts in the training data, devoid of any real-world meaning or design intent. This mirrors the potential for LLMs to generate plausible-sounding but ultimately nonsensical outputs based on patterns in their training data. This also relates to the critique of purely data-driven views.

NotebookLM continued: These examples illustrate how the framework of the previous explanation categories can be used to generate absurd and nonsensical justifications for a design decision. They often rely on flawed logic, irrelevant rules or precedents, bizarre metaphors, or spurious data correlations.

Kudos to the LLM for constructing these colourful illustrations of absurd reasoning. That said, whether sourced from flawed logic, arbitrary rules, irrelevant precedents, uncomfortable metaphors or spurious data, absurdities have their place in design processes and outcomes. See post Nonsensical signs.

Note

• Featured image is by Dall-e, “featuring a vast murmuration of starlings swirling above and around a decaying apartment block, creating a striking contrast with the urban decay.”