G2: Liberty versus Safety

Four quick rulings

Imagine you sit in a legislature, and four proposals arrive on the same afternoon. Answer each one quickly, yes or no, before reading on.

First: social media companies must check users' ages and keep children under sixteen off their platforms.

Second: an adult must pass a background check and wait several days before buying a gun.

Third: the police may read a suspect's encrypted messages, with a warrant, because criminals use encryption to hide.

Fourth: a bar owner may let customers smoke indoors, because nobody is forced to drink there.

Now look at the reasons you gave yourself. On some of these you probably reached for freedom: the right of adults, parents or owners to make their own choices, and the danger of letting the state decide for them. On others you probably reached for safety: the harm prevented, the people who cannot protect themselves, the cost of doing nothing. Few people answer all four for the same reason. Most find that freedom decided some and safety decided others, and that which one it was had a good deal to do with which side of politics the proposal seemed to come from.

That is not necessarily hypocrisy; weighing two real values differently in different cases is what a thoughtful person should do. The trouble starts when each ruling is defended as though only one value existed.

What the grammar is

Almost every rule that restricts what people may do is defended as protecting someone from harm, and almost every protection from harm restricts what someone may do. Freedom of action and protection from harm are both real goods, and in many cases more of one means less of the other. That is a genuine trade-off. This grammar is what happens when it is argued as though it were not one: as though only one of the two values were real, and the people who weigh the other were fools or tyrants.

The deep structure has a second layer, which the four rulings were meant to expose. Because most people's weighting of the two values flips with the topic, the principle a person states is often not the one actually deciding their position. Someone who says "people should be free to take their own risks" about one thing may say "we cannot allow this danger" about another, with no account of what makes the cases different. The stated principle is sincerely felt, but something else is doing the choosing.

The grammar also has a favorite way of forecasting. Slippery-slope reasoning, the claim that a first step will lead by stages to something much worse, runs through it in both directions. One side treats every new rule as the first step toward tyranny; the other treats every repealed rule as the first step toward chaos. Both skip the question that would make the forecast useful: what, in this case, would actually stop the slide?

This site calls a recurring pattern of argument like this a grammar: a shape the same mistake takes across many different subjects, so that once you can see it in one argument you can see it in others. This one has two poles, and each is a real failure.

Pole A, the liberty absolutists, ignore real externalities in the name of freedom. An externality is a cost that falls on someone who did not choose it: the smoke a non-smoker breathes, the injury to a bystander, the bill that lands on the public when a private risk goes wrong. The liberty absolutist talks as if the choice in question affected only the person making it. Sometimes that is true. Often it is not, and the argument has no place for the people who bear the cost.

Pole B, the safety maximalists, ignore the real costs of a protective rule in the name of safety, and treat any conceivable harm, however small, as enough to justify one. If something could hurt someone, that is enough to restrict it, and the costs of the restriction itself, in lost freedom, lost opportunity or new risks, go uncounted. The safety maximalist hears any objection as indifference to the victims, and talks as if a rule's good intentions settled whether it works or is worth what it takes away.

Each pole counts one column of the ledger and declares the other column empty.

Taking the trade-off apart

Undoing the grammar starts with refusing to let either value disappear. A few questions do most of the work.

Whose freedom, and whose safety? These arguments are rarely about one person's liberty against the same person's safety. More often one group's freedom is weighed against another group's protection, and naming the groups changes the argument. A rule about what adults may do to themselves is a different case from a rule about what they may do near others.

How large is the harm, how likely, and how large the cost? "Could cause harm" is true of nearly everything. The useful question is how much harm, to how many people, compared with what would happen otherwise, and the same question applies to the freedom a rule removes. A restriction can be justified in principle and still fail in practice; a freedom can be valuable in principle and still be used mainly to push costs onto others.

What would stop the slide? When someone argues that a first step leads to a worse one, ask what lies between them: laws, courts, elections, public opinion, the plain difficulty of going further. Sometimes nothing does, and the warning is sound. Sometimes a great deal does, and the slope is a rhetorical device.

Would I say the same about the other case? Take your reasoning on one topic and apply it, word for word, to a topic where you hold the opposite view. If the reasoning still works there and you reject the conclusion, you have found where something other than your stated principle is deciding.

Asked together, these questions turn a contest between freedom and safety into a narrower disagreement about a particular harm, rule and group of people.

Why it is so persistent

The simplest reason is that "this is about freedom" and "this is about keeping people safe" are slogans, while "this is a trade-off, and here is how I would weigh it" is a paragraph.

A second reason is the gap between stated and revealed principles. When a person's weighting of liberty and safety shifts with the topic, the shift is invisible from the inside. Each position feels principled on its own; only lining them up side by side shows that the principle changed between them.

Source still being found. research on whether people's support for liberty or safety arguments shifts with the political identity of the group affected, for example studies of motivated reasoning applied to civil liberties

A third reason is how people perceive risk. Risk literacy, the ability to compare dangers by how likely and how large they are, is hard, and people judge risks partly by other features: whether a risk is chosen or imposed, familiar or new, vivid or abstract. That makes it easy to feel that one's own freedoms carry acceptable risks and other people's carry unacceptable ones.

Source still being found. the research literature on risk perception showing that chosen, familiar or controllable risks are judged less dangerous than imposed or unfamiliar ones of similar size

Finally, each side can draw on real history. The liberty side can point to rules sold as protection that became tools of control; the safety side can point to freedoms defended as principle while people were hurt. Each remembers its own cases and forgets the other's. The precautionary principle, the idea that serious potential harm justifies caution before the evidence is complete, shows the pattern: a sensible rule in some settings, easily stretched into a reason to restrict anything at all.

Source still being found. a reliable account of the origin and standard formulations of the precautionary principle, and of the main criticisms of how it has been applied

How the argument has been had before

Each of these earlier fights was argued as freedom against safety, and in each both sides were right about something. The details are still being checked, so what follows sketches the shape of each argument rather than its history.

Seatbelt laws. When governments began to require seatbelts, one side argued that belts reduce deaths and injuries; the other, that adults should decide what risks to take with their own bodies. The liberty side was right that the main risk of going unbelted falls on the person who chooses it. The safety side was right that it does not fall only on them, since others share the costs of injuries. The question that mattered was how large those shared costs were, and neither slogan answered it.

Source still being found. when and where mandatory seatbelt laws were introduced, a quoted contemporary argument from each side, and what the evidence showed about their effects

Smoking bans. Bans on smoking in workplaces, bars and restaurants turned on an externality that was easy to see and hard to measure: other people's smoke. The liberty side was right that owners and customers were making choices. The safety side was right that the people who worked there had not all chosen the smoke. The real question was the size of that harm.

Source still being found. dates and places of major indoor smoking bans, quoted arguments from both sides, and the state of the evidence on secondhand smoke when the bans were debated

Drug prohibition. Here both poles have the longest record. The safety side has argued that some drugs harm users, families and neighborhoods so badly that banning them is justified. The liberty side has argued that adults may decide what to put in their bodies, and that prohibition creates harms of its own, such as black markets and the violence that comes with them. Each side's strongest point is the other's blind spot: the costs of the drugs, and the costs of the ban.

Example still being found. a quoted, dated historical argument for a prohibition regime (for example US alcohol prohibition, 1920–1933, or later drug laws) that ignored the costs of the ban, and one against prohibition that ignored the harms of the drug

Surveillance after 11 September 2001. After the attacks, the United States passed the USA PATRIOT Act and expanded government surveillance. Supporters said new powers were needed to prevent further attacks; opponents said the powers threatened privacy and invited abuse. Slippery slopes were argued openly on both sides: refuse the powers and more attacks would follow; grant them and a surveillance state would follow. Which forecast came closer is a factual question that deserves a documented answer, not a remembered one.

Source still being found. the main surveillance measures adopted after 9/11, how they were used, and what later disclosures or official reviews found about their effectiveness and abuse

Example still being found. a quoted, dated argument from the period that treated security as the only value at stake, and one that treated privacy as the only value at stake

COVID-19 mandates. During the pandemic, governments and institutions imposed lockdowns and mask and vaccine requirements. The safety side argued that infection is an externality, since one person's choice affects others' risk. The liberty side argued that the measures were broad, long-lasting and left little room for individual judgment. Both were arguing about a real trade-off under real uncertainty, and both often spoke as though the other value did not exist.

Example still being found. quoted, dated statements from the pandemic period showing each pole, from public officials or prominent commentators on different political sides, with a non-partisan overview of the evidence on the mandates' costs and benefits

In each, a clash of values turned largely on questions of size: how large the harm, how large the cost, and who bore each.

How each pole fails

The two poles fail in mirror-image ways.

The liberty absolutist treats a choice as private when it is not. The cost that falls on someone else is left out, or called that person's problem, or answered with a principle ("it's my right") where a measurement was needed. The slope runs one way only: every rule is the first step toward tyranny, and what would stop the slide is never asked. And the principle is often selective, defending the freedoms the absolutist values while accepting restrictions on freedoms they do not.

Example still being found. a real, quoted, dated present-day instance of Pole A (liberty absolutism that ignores a real externality) from the political right

Example still being found. a real, quoted, dated present-day instance of Pole A from the political left, or from a civil-liberties position not usually coded right

The safety maximalist makes the opposite moves. A possible harm is treated as sufficient reason, without asking how likely or how large. The costs of the rule go uncounted: what it takes from the people restricted, what it fails to prevent, what new harms it creates. The slope runs the other way: every loosened rule is the first step toward chaos. And the caution is often selective too, demanding protection from the dangers the maximalist fears while defending freedoms whose dangers they discount.

Example still being found. a real, quoted, dated present-day instance of Pole B (safety maximalism that ignores the costs of a rule) from the political left

Example still being found. a real, quoted, dated present-day instance of Pole B from the political right

Both poles take a case with two real values and argue as if it had one. And because most people stand at Pole A on some topics and Pole B on others, the move belongs to no single side.

Where it borders other grammars

With G3: Sacred Values versus Trade-offs it shares the refusal to weigh. When "children's safety" or "free speech" is invoked to end an argument rather than begin one, a value has been declared beyond price, and the trade-off can no longer be discussed.

With G7: Moral Panic and the Folk Devil it shares slippery-slope forecasting and alarm that outruns the scale of the harm. Many safety campaigns begin as a panic, and many liberty campaigns begin as a panic about the panic.

With G9: Language as Territory it shares the fight over names. Whether a rule is called "protection" or "censorship," "public health" or "control," "moderation" or "suppression," often decides which value the listener hears before any argument is made.

When naming the pattern becomes a weapon

From the liberty side, labels such as "nanny state," "authoritarian" or "fearmongering" can dismiss a real harm without measuring it. Calling a worry a "slippery slope fallacy" can do the same work: some slopes really are slid down, and calling a forecast a fallacy does not show that this one is wrong. From the safety side, labels such as "reckless," "selfish" or "anti-science" can dismiss a real cost without counting it, and cast anyone who raises a rule's cost as indifferent to its victims.

Example still being found. a real, quoted, dated instance of "nanny state" or a similar label used to dismiss a measurable harm, and a real, quoted, dated instance of "selfish" or "anti-science" used to dismiss a measurable cost

This page's own vocabulary can be turned the same way. "You only care about liberty when it suits you" may be true, and it may also be a way to avoid the argument. Showing that someone's principles shift with the topic shows that their stated principle is not doing the work; it does not show that their position is wrong in this case. That is the weaponized form of A4: The Mirror Check: explaining away a belief by diagnosing the believer's motives instead of answering the claim. Calling something "a trade-off" can be a dodge too, sounding balanced while declining to say how one would weigh it.

Catching it in yourself

Go back to the four rulings at the top of this page.

Where did I reach for freedom, and where for safety? List the topics on which freedom decided your view and those on which safety did. If the lists line up neatly with your politics, the grammar is at work in you.

What would I say if the other side proposed this? Imagine the same rule, with the same evidence, coming from people you usually oppose. If your answer changes, your principle was not deciding it.

Whose costs am I not counting? Where I favor freedom, who bears the risk I am defending? Where I favor safety, who bears the cost of the rule?

What would stop the slide? When I warn that a step leads somewhere worse, can I say what stands in the way and why it would fail? When I dismiss someone else's warning, can I say why it would hold?

What would change my mind? What size of harm would make me accept this restriction, and what size of cost would make me drop it? If no answer is possible, the value has become sacred, and that is a different grammar: G3.

Tools that help

• B3: The Trade-off Surfacing. Ask "compared to what, and at the cost of what?" This grammar works by hiding one side of a trade-off, and this tool puts the freedom given up and the harm accepted in the same sentence. Its misuse is treating every firm commitment as a mistake merely because it has a cost.
• F1: The Crux Finder. Locate the specific upstream question the disagreement depends on. Most of these fights turn on a factual question (how large the harm, whether the rule works) and a question of weighting (how much freedom for how much safety, and for whom); finding which is actually in dispute makes the argument answerable.
• A3: The Newsfeed Inversion. How often something appears in your feed measures how well it performs, not how common it is. Safety arguments are often driven by vivid, widely shared cases of harm, and liberty arguments by vivid stories of overreach; checking how often each actually happens keeps either from setting the terms alone.

Four tools work against every grammar on this site: A4, C2: The Steelman Protocol, E5: The Precision Covenant and F4: The Scout Report. C2 is especially useful here: state the other side's value so well they would sign it before arguing against their weighting of it. See all thirty tools, or what this site means by a tool.

Where it shows up now

As of October 2026, the site has not yet examined this grammar in a live argument. These are places to look, not findings.

• Children's phones and social media bans (education). Not yet examined.
• Gun policy (liberty). Not yet examined.
• Content moderation and "jawboning," meaning informal government pressure on platforms to remove content (media, liberty). Not yet examined.
• AI regulation (AI). Not yet examined.
• Helmet-and-playground culture, the arguments over how much risk children should be allowed to take (education). Not yet examined.
• Encryption backdoors, proposals to give governments a way to read encrypted communications (liberty). Not yet examined.

Related concepts

• Revealed versus stated principles. The difference between the principle a person says guides their views and the principle their views across many cases actually reflect.
• Trade-off denial. Refusing to acknowledge that a choice gives up something of value.
• Risk literacy. The ability to understand and compare risks by how likely and how large they are.
• The precautionary principle and its abuses. The idea that a serious potential harm justifies caution before the evidence is complete; its abuse is invoking it against any conceivable harm, or only against harms one already wanted to restrict.
• Externality. A cost or benefit of a choice that falls on someone who did not make it.
• Slippery-slope reasoning. The argument that a first step will lead by stages to a much larger result; sound when something pushes along the slope and nothing stops it, unsound when neither is shown.