Norms Are Broken Laws

sirawit4 Aug 2026 23:42 UTC
−14 points
5 comments4 min readLW link

An Introductory Example

Consider the following unwritten rule:

Keep the kitchen clean. Clean after you cook.

Consider the following three characters: the unreflective individual, the conscious egoist, and the rational moralist.[1]

  1. The unreflective individual cleans the kitchen after cooking, almost automatically, because it is a norm that the society expects people to comply with. The only cost is the physical cost to clean the kitchen. The perceived cost is none, since that’s the default thing to do regardless of others.

  2. The conscious egoist doesn’t clean the kitchen. The conscious egoist examines the dilemma whether to (A) clean the kitchen or (B) don’t clean the kitchen. The conscious egoist measures the cost of (A) as the physical cost, and the cost of (B) as none, so the conscious egoist chooses (B). The final perceived cost is none.

  3. The rational moralist faces the dilemma (A)/​(B) as above. The rational moralist measures the cost of (A) as the physical cost, the cost of (B) as the emotional cost (i.e. I’m unhappy that I’m a bad person acting as a free-rider). If the rational moralist chooses (A), then they bear the physical cost. If the rational moralist chooses (B), then they bear the emotional cost.

I’ll generalize the kitchen example to “all [unwritten but expected] norms”. Let me introduce more examples here:

  • Keep quiet on public transport.

  • In the school cafeteria, return the trays, don’t just leave them on the table.

  • Be polite in communications.

  • Respect others.

  • Keep your promises.

If you replace the kitchen scenario with other examples, you can see that the cost analysis of three characters above remain the same, except for the fifth example, where non-compliance carries a significant cost.[2]

My Claim

There is an asymmetry whenever the society expects its members to follow some unwritten norms. My main claim is that:

All (unwritten but expected) norms push the burden into the rational moralist, creating a perceived tax (whether physical or emotional cost). This punishes the rational moralist, and therefore such society encourages people to rather be unreflective or egoistic, compared to being rational and moral. The collective judges efficacy competently and distribution blindly[3], hence leaving things as they are, not accepting the decision to convert norms into written laws.

My proposal is therefore to minimize [unwritten but expected] norms, i.e., “abolish all norms”. This doesn’t directly mean actively removing all the norms, but rather, I mean: Revise the norms, turn them into a written rule that is agreed in the social contract, if verifiable, sanctionable, and if the failure of its compliance leads to a negative social outcome. For other norms (such as “saying thank you”, “voluntary kindness”, etc.), move them towards the etiquette category, and do not expect people to follow such norm, and also discourage people to expect others to follow such norm.

Under this revision, the unreflective individual continues to perform whatever has to be done normally, just that the unwritten norm “Keep the kitchen clean. Clean after you cook.” is turned into an explicit rule:

Members of the kitchen must clean after cooking. Failure to do so will result in a fine of $100.

Now the conscious egoist and the rational moralist both face a cost: choosing (A) creates a physical cost, and choosing (B) creates a concrete $100 cost.

This new protocol distributes the old “rational moralist tax” into a universal tax: one either cleans the kitchen or pays the fee. I argue that this system is fairer than the original one.[4] Now let us abstract this example away and let me propose a full formal specification below.

The Blueprint

Let be an [unwritten but expected] norm. Let be the perceived cost of complying with the norm. Let be the perceived cost of not complying with the norm. In the original situation, the unreflective individual pays zero cost. The conscious egoist pays zero cost. The rational moralist pays .

Legalizing blueprint: Choose an appropriate[5] cost to apply universally. Impose a new rule:

Every member of this society must either obey or pay .

Now, everyone pays (although the unreflective individual might interpret this as zero since obeying is the default choice which was free from the beginning), assuming that the moralist perceives “not complying to the norm” as “paying the fee and retaining no extra moral obligation”.

Law, Norm, and Etiquette

There are at least four necessary conditions for something to be a law:

  1. It must be an enforcement, not a description, not an advice.

  2. The enforcement must apply to humans, not animals, not non-living things.

  3. The enforcement must enforce actions, not thoughts, and such actions must be verifiable.

  4. It has a sanction: there is a penalty for failure of compliance.

  5. It must be enforced to all members of the society and must be agreed in the social contract.

On the other hand, norms are unwritten rules, but the society expects everyone to comply. Etiquette is a weaker version of norms: the society doesn’t really expect everyone to comply, but complying would signal a positive social value. Laws, norms, and etiquette form a basic organization instrument for a society. This has happened historically, and can be seen as “rules that shape the social coordination”.

As I have argued above, my belief is that something that is not enforced but is expected always creates an asymmetric cost to some particularly conscious agents (in this example, the rational moralist). That is,

A norm is merely a law enforced only against the conscientious.

A norm’s enforcement mechanism is conscience, and conscience only reaches agents who have one. Therefore, a norm satisfies the above conditions 1-4 while failing the 5th one. Society tries to enforce, but not against everyone.

I accept that norms and etiquette are important ingredients of the current civilization, but I also argue that this is not the best way. I believe that the blueprint above would turn enforceable norms into laws and either discard the unenforceable ones or move some towards etiquette. One might argue that a historical reason of norms is that it is a cheaper alternative to laws, since a formal prosecution of a routine behavior might push everything into sluggish bureaucracy, and that’s not worth the price. My claim is that I believe it’s worth the price, not in the sense that everyone will be better off, but in the sense that I think this move would fix the fairness problem while not reducing the societal benefit by much.

Therefore,

Norms Are Broken Laws.

Those worth enforcing should be enforced on everyone. Those not worth enforcing should not be expected of anyone.

“Expecting people to comply” creates an asymmetric tax on the rational moralist, and therefore conversion to written laws with active enforcement would be a fairer alternative.[6]

  1. ^

    I’m not claiming that everyone belongs to one among the three characters. There may be others, but I’m not considering the other types here, for the reason of simplification. I argue that it is enough to consider only these.

  2. ^

    Although they form different conclusions. For instance, if one were to run “respect others” through the blueprint, one sees that the corresponding law must at least distinguish between having respect and not having respect, which is not actually verifiable. On “keep your promises”, the current unwritten sanction already did the job well, so no law codification is needed. If the egoist breaks a promise, then the society knows that the egoist breaks, removing reputation from the egoist, and this is already a cost. Therefore this analysis doesn’t match the template as the other four. The other three examples have a concrete objective witness, hence verifiable. They are also enforceable by law. “Be polite in communications” is weak, since it raises the question “from objective evidence, what counts as polite?” My response would be to break down “politeness” into verifiable pieces: “Do not use slurs”, “Do not make threats”, etc.

  3. ^

    The collective can observe whether the kitchen gets cleaned. It cannot observe who paid for it.

  4. ^

    Some people argue that this decision turns the punishment structure into economic gamification, and the result may be worse (i.e. no one cleans the kitchen while everyone pays the fee). My response would be “then increase the fee so that the marginal cost/​benefit of cleaning the kitchen becomes better than paying the fee”.

  5. ^

    Ideally, pick , for some small (procedural) , so that the choice to comply is a bit better than not complying. Note that this design choice might unintentionally increase the cost to the rational moralist (when , for instance), so the blueprint may not work in favor of the rational moralist but rather redistribute the moral cost to the egoist. Solving the fairness problem doesn’t imply the benefit/​cost problem being solved.

  6. ^

    Note that I’m claiming very weakly about the public benefit. I’m claiming only that this blueprint doesn’t create much public cost compared to the public benefit. This is a belief, and one can convince me otherwise. Perhaps in a more careful analysis, one might consider the public benefit arising from one compliance of the norm and the public cost arising from one non-compliance of the norm, if one really wants to argue (in a utilitarian way) about a public protocol. This analysis mainly considers the fairness part, and is trying to do justice to the rational moralist.

sirawit4 Aug 2026 23:42 UTC
−14 points
5 comments4 min readLW link