In a recent post, I outlined some problems I identified with Scott Alexander’s Noncentral Fallacy. He identified a very real problem in the discourse, but I disagreed with him about what made it fallacious. Briefly, Scott believed the problem lay in the use of applying technically correct but non-central labels to evoke misleading connotations about category membership (e.g., while MLK is technically a “criminal”, he’s not the typical archetype of a criminal). However, I argue that the real issue is not the centrality of the label but the use of connotation-heavy labels as a way to shortcut substantive debate (e.g., calling abortion “murder” to circuitously imply it is or should be unlawful). This tactic, which I dubbed the Sticker Shortcut Fallacy, manifests in fixating on categorization membership as a way to bypass substantive argumentation and potentially secure agreement over contentious premises without proper justification.
The most interesting responses to my post were the ones that...reenacted the exact fallacy I described while insisting it was not a fallacy! To be clear, I don’t believe that any of those responses were motivated by malicious or sophistic intent. Rather, I believe it’s an illustration of how pervasive and insidious this fallacy is. This follow-up aims to precisely explain why the Sticker Shortcut Fallacy is indeed the result of fallacious reasoning and how vulnerable we are to it.
I’m at the combination composition and division
The best way to approach this fallacy might be to understand it as a combination of composition and division fallacies. To recap, the composition fallacy occurs when one assumes that what is true of the parts must be true of the whole (if each ingredient is tasty, then the finished dish must also be tasty). Conversely, the division fallacy assumes that what is true of the whole must be true of the parts (if the meal is tasty, then each individual ingredient must also be tasty).
The Sticker Shortcut Fallacy leverages both by applying it to semantic conceptual categories. Instead of parts within sums, we’re talking about things within categories. The problem occurs when a trait commonly found among category members is reified as a necessary or “central” requirement of the category constellation (composition fallacy). Then, you return to earth and proclaim that any putative member lacking this newly declared essential trait is excluded from the category (division fallacy).
For example, observing that most birds can fly, assuming that flight is a necessary trait for being classified as a bird (composition), and subsequently excluding penguins from being birds because they don’t fly (division). Or observing that mammals tend to have fur, assuming fur is a necessary trait for being a mammal, and therefore excluding dolphins as not mammals. Or observing that weapons tend to cause bleeding and therefore excluding blunt instruments like clubs and batons. The list goes on.
The fallacy should be apparent when laid out in such simple terms. However, the key reason for its confusion and continued pervasiveness is that people don’t typically sit down before a conversation to agree on the specific contours and definitions of the vocabulary, terminology, or labels they intend to use. Practically speaking, it’s not reasonable to expect detailed definitions of every word in everyday conversations — we necessarily rely on shared assumptions and general understandings when we communicate.
The real issue lies in the inherent ambiguity and fluidity of language. Labels and words are used by individuals (reasonably) assuming that their listeners share their understanding of a term, and so misunderstandings over meaning often are innocuous. However, you should absolutely be suspicious of anyone who demonstrates an obsession with enshrining specific labels and terminology and refuses to explain why. This behavior is the essence of the Sticker Shortcut Fallacy — manipulating conversations by surreptitiously imposing their own definitions as a vehicle to sneak in their highly contested premises.
Only the lawyers can save us now
The Sticker Shortcut Fallacy is not possible without the crucial ingredient of linguistic ambiguity. If we agreed ahead of time on a strict definition of the weapon semantic category — say, any device used to inflict physical harm — we’d very likely agree on what to accept into the category (knives, guns, howitzers, space lasers, pebbles, etc.). We might notice some other common trait within our corral — say, most weapons tend to cause bleeding — but we could exclude something like a baton from the weapon category only if we agreed to change our membership criteria (Related: Disguised Queries).
Lawyers are notoriously allergic to linguistic ambiguity, as evidenced by the sheer cliffs of verbosity present within dense and convoluted legislation and statutory codes. Typically, the legal system gives no fucks about making the law legible to plebs, but there’s a rare exception for jury instructions. Capitulating to practical necessity, jury instructions showcase the possibility of distilling weighty tomes into succinct guidance that still maintains sharply-honed legal precision. I find the depths of granularity they plumb down to fascinating.
Suppose someone is on trial accused of “rape”. You’d would be careening towards inevitable catastrophe if you burdened 12 random people with the responsibility of adjudicating that person’s guilt without precisely defining the offense. Colloquially, “rape” can describe a wide range of non-consensual sexual activities, including any form of unwanted touching. In contrast, as a random example, the North Carolina pattern jury instructions for Second Degree Rape lucidly and unambiguously spells out vaginal intercourse as one of its elements, which in turn is defined as penetration of the female sex organ (however slight) by the male sex organ.[1] There’s plenty to disagree about regarding the application of this element to any particular scenario but, crucially, the ambient amount of ambiguity and corresponding misunderstanding is drastically eliminated.[2]
Keep your assumptions pared down
Remember, our semantic categories are products of arbitrary human folly and can be obscenely misleading. Sure, at their apex, they help us sort through the world and make plausible predictions about novel phenomena. If we encounter a creature with dry, scaly skin basking in the sun, we can likely deduce that it lays eggs and is cold-blooded, even if we’ve never seen it before.
But if you’ve agreed on a precise delineation of a category’s boundaries, then the only information you should glean from finding out “X is a weapon” should be “X is a device used to inflict physical harm,” or whatever else you explicitly define about the category.
However, words are potent weapons (see what I did there?) and unless you’re a consistently pedantic wordcel, it’s extremely difficult to jettison a vocabulary’s baggage completely. This is especially true for culturally load-bearing words that hold up significant emotional salience. Humans might be savvy pattern-matching machines but also incorrigible social apes. It’s really easy to lose track of the membership criteria we might have previously ratified and instead sneak in our own assumptions or absorb them from our ambient surroundings.
Referring to someone as a “convicted criminal” inevitably conjures up all sorts of negative connotations. However, this sticker label, on its own and narrowly understood, actually tells you nothing about whether the individual did anything wrong, illegal, or immoral. If you commit to keeping your assumptions pared down, the only information the “convicted criminal” label should convey is “the pertinent legal system has deemed the respective individual guilty of a criminal allegation”. Nothing more. Of course, it’s perfectly reasonable to adopt a host of plausible Bayesian predictions about said person, but you must always remember that you are spelunking within the assumptions cave. Downstream connotations should be viewed as probabilistic rather than determinative.
With any categorical skirmish, avoid falling into the assumption pit and instead ask why the category matters. Maybe the new hire forgot that we started sorting bleggs from rubes as an imperfect-but-good-enough method to efficiently prospect for palladium and vanadium ore. Misunderstandings are often innocuous and those are easily resolved by clarifying our baked-in assumptions.
All language is made-up, but it remains useful only to the extent others share your understanding.
If you want to communicate X, just communicate X! Don’t rely on category Y to smuggle in X as a connotation stowaway.
Misunderstandings are common and expected, which is why definitions are crucial. Always start by ensuring you have shared definitions to avoid confusion. It’s still possible that someone insisting on speaking indirectly through ambiguous semantic clouds does so without malicious intent, possibly because they’ve never understood the concept any other way.
However, an obsession with specific labels that refuses to clarify their importance is unlikely to be anything but malicious. Stay safe out there and avoid this trap.
- ^
No, this does not mean that North Carolina only punishes heterosexual coercive acts. The state has a separate statute criminalizing any coerced “sexual act” that it punishes with the exact same severity. Still, legally, it refers to it as “sexual offense” rather than “rape.”
- ^
It’s impractical to eliminate ambiguity entirely because adding to the pile of verbose legal jargon very quickly hits diminishing returns. This is evidenced by the fact that many people today continue to make lots of money based entirely on their ability to have pedantic lawfights about vocabulary. In this arena, judges resort to an array of specific canons of interpretation (such as legislative intent, ordinary meaning, historical meaning, rule of lenity, etc.) grounded in legal precedent to resolve vocabulary disputes. That’s why bees are fish, of course.
I’d appreciate examples of the sticker shortcut fallacy with in-depth analysis of why they’re wrong and how the information should have been communicated instead.
I wanted to include very basic examples first:
I am planning yet another follow-up to outline more contentious examples. Basically, almost any dispute that is based on a disguised query and hinges on specific categorization matches the fallacy. Some of the prominent examples that come to mind, with the sticker shortcut label italicized:
Was January 6th an insurrection?
Is Israel committing a genocide?
Are IQ tests a form of eugenics?
All of these questions appear to be a disguised query into asking whether X is a “really bad thing”. But instead of asking this directly, they try to sneak in the connotation through the label. Similarly, the whole debate over whether transwomen are women is a hodgepodge of disguised queries that try to sneak in a preferred answer through the acceptance of labels. In each of these examples, we’re better served by discussing the thing directly rather than debating over labels. Does this help clarify?