Appeal to force

“Explain it at your review”

At a planning meeting, two engineers say the launch date is unrealistic. The project director replies: “The schedule is realistic. And anyone who still thinks otherwise is welcome to explain that at their performance review.”

The director has given a reason to stop saying the schedule is unrealistic, and no reason to think it isn’t. Whether the work fits in the time depends on the work. This is the clear-cut form: the threat is aimed at a factual question, and its effect is to shut down the people best placed to answer it. The Stanford Encyclopedia of Philosophy notes that in the pragma-dialectical theory of argument, this kind of intimidation is a fallacy because it breaks the rule that you may not try to prevent the other party from expressing their views.

A threat dressed as advice

A new member of a community garden asks at the annual meeting how the plot fees were spent. Afterward, a longtime board member takes her aside: “Just some friendly advice. People who make a fuss about the accounts somehow always end up with the plot next to the compost heap. I’d trust that the numbers are fine.”

On the surface this is a warning, even a kindness. But the board member helps assign the plots, so the “advice” is a threat from the person who would carry it out. Douglas Walton analyzes a case with the same shape, a gangster “advising” a shop owner that the neighborhood is dangerous for people who don’t pay. On his account the fallacy works by pretending to advise: the threat is left implicit so that the speaker can deny having made one. Nothing about the fees has been shown either way.

Believing under pressure

A cycling magazine is about to run a harsh but carefully tested review of a new bike light. The light’s maker, the magazine’s biggest advertiser, mentions that its ad contract is up for renewal next month. The editor rereads the review and concludes: “On reflection, the light is actually pretty good. The review is too negative.”

No explicit threat was made, and the editor might be excused for thinking about the ad contract when deciding what to publish. The error is in the conclusion about the light: the tests haven’t changed, and the renewal date is no evidence about how bright it is. The Internet Encyclopedia of Philosophy makes this point about a similar case: the one applying pressure uses a scare tactic, but it’s the person who accepts the claim because of the pressure who reasons fallaciously.

The right-hand column is the scheme Walton and colleagues call argument from threat. Its conclusion is about what it would be prudent to do, and whether it’s a good argument depends on whether the threat is real and what complying would cost. It is not fallacious in itself: Walton’s starting example is a robot on Mars that threatens to break another robot’s antenna unless it digs, and the second robot has good practical reason to weigh the threat. The left-hand column uses the same kind of pressure to settle a question of fact, and there it has no bearing at all.

• Direct threat: “agree, or else.”
• Veiled threat: the consequence is described as something that just tends to happen, while both sides understand who would make it happen. Walton notes that indirect threats like this have been harder to pin down and evaluate than direct ones.
• Pressure the target applies to themselves: no one threatens, but a belief is adopted because disagreeing would be costly, as in the magazine example.

What separates a threat from a warning? In Walton’s analysis of the two speech acts, a warning tells you about a bad outcome that you can avoid. A threat adds one condition: the speaker commits to bringing about the bad outcome unless you do what they want. If that commitment is absent, it’s a warning.

Is it a kind of appeal to consequences? Walton analyzes threat and fear appeals as sharing the underlying structure of an argument from negative consequences, with the difference that in a threat the speaker is the one who will cause the consequences. The pragma-dialectical approach, as described by Francisca Snoeck Henkemans, instead treats a threat appeal, an appeal to fear and a fallacious appeal to consequences as violations of three different rules. Because the classifications disagree, this site lists them as related rather than filing one under the other.

Is it an argument at all? Here too the literature is divided. Walton, following John Woods, treats even a direct threat as a prudential argument (an argument about what it would be wise to do). Don Levi argues that the ad baculum is not a fallacy in an argument but something offered instead of argument, to put an end to discussion, and John Deigh holds that a robber pointing a gun is no more making an argument than a growling dog is. Alan Brinton, against Michael Wreen’s view that it need not, argues that an ad baculum essentially involves a threat made by one person to another.

Force, fear and emotion. A threat usually frightens, which is why some references cross-refer ad baculum to the appeal to emotion (fear). Walton’s Scare Tactics (2000) keeps them apart: an appeal to fear describes a frightening outcome without threatening it, and a threat need not be meant to frighten. Francisca Snoeck Henkemans, reviewing the book, calls that distinction ill-motivated, pointing out that the purpose of a threat can perfectly well be to frighten.

Introductory textbooks such as Hurley and Watson’s define the fallacy by its target: a threat of harm if the listener doesn’t accept the conclusion, which is always irrelevant to whether the conclusion is true. Once the conclusion is about action instead, things are less clear. The Stanford Encyclopedia of Philosophy calls the ad baculum “one of the most controversial” fallacies, “because it is hard to see that it is a fallacy or even that it involves bad reasoning.” Threats and consequences are often legitimate parts of reasoning about what to do.

• It’s a warning, not a threat. Pointing out bad consequences that the speaker doesn’t control is ordinary advice.
• The conclusion is about what to do. Threats give real reasons to act. Charles Kielkopf’s example: threatening drivers with the loss of their license is a perfectly reasonable way to get them to accept that they shouldn’t drive drunk, read as a call to act.
• The setting is a negotiation. Walton notes that making threats is normal in negotiation, such as a union threatening to strike during contract talks. The Stanford Encyclopedia of Philosophy gives strikes and cutting off trade routes as examples not normally considered fallacies.
• Stating a rule and its penalty. “Books returned late are fined a dollar a day” announces a consequence to guide behavior; it doesn’t ask anyone to believe anything because of it.
• Complying while still disagreeing. Going along with a demand because the cost of refusing is too high can be prudent. It becomes the fallacy only if you conclude that the demand was right because of the pressure.

According to Walton, whether a threat is fallacious depends on the kind of discussion it’s part of: it is simply out of place in a discussion aimed at finding out what’s true, but can be normal in bargaining.

The test: is the threat offered as a reason to do something, or as a reason to believe something (or to stop questioning it)?

The fire hydrant

A neighbor says: “If you keep parking in front of that hydrant, you’re going to get towed. They took my brother’s car from that exact spot last month, and it cost him three hundred dollars.”

This describes a bad consequence to get someone to change their behavior, which is the surface of an appeal to force. But the neighbor doesn’t control the tow truck, so it’s a warning rather than a threat, and the conclusion is about where to park, not about what to believe. It meets the warning and about what to do conditions.

The overdue invoice

A freelance designer emails a client: “The invoice for the first phase is now six weeks overdue. If it isn’t paid by Friday, I’ll pause work on the second phase until it is.”

This is a genuine threat: the designer will bring about the consequence. But it’s made in a business relationship where each side’s leverage is a normal part of the terms, and it asks the client to do something, not to believe something. Nothing is being settled by intimidation that should have been settled by evidence. This is the negotiation and about what to do conditions.

Threats work. As a way of changing what people do, they’re often faster and more reliable than argument, and someone with power over others can use them without having an answer to the objection at all. Hurley and Watson add that the threat usually works by keeping the listener from noticing a hidden premise that, stated out loud, is plainly false: that if someone can be forced to admit a claim, the claim is true. A threat is good evidence that someone might get hurt, and it’s easy to let that stand in for evidence about the claim. Walton points out that when a threat is veiled, the speaker gets its effect while keeping the ability to deny having made it.

It’s also easy to err in the other direction. Because threats are unpleasant and sometimes illegal, Walton notes, it’s tempting to label any argument that mentions a threat as fallacious, even though threats are routine and legitimate in settings like bargaining. On the receiving end, people under pressure may find it easier to change their minds than to go along with something they still think is wrong, which is how a threat that only justified compliance ends up producing belief.

• Separate what you’ll do from what you think. “I’ll work to the schedule, but I still think it’s unrealistic, and here’s why” keeps both on the record.
• Ask who controls the consequence. If it’s the person telling you about it, treat the “advice” as a threat, and look for the argument that should have been given instead.
• Notice when a question has stopped being discussed. A threat often shows up as silence: the objection simply isn’t raised again.
• If you hold power over others, watch for disagreement you’ve made costly without meaning to. The people who stop objecting may be the ones who know the answer.