How To Believably Write Serious Mistakes

As I’ve mentioned numerous times before, I’m not a huge fan of AI technology. So it was perhaps with an inappropriate amount of entertained interest that I watched the meltdown of OpenAI at the end of November. To sum up a lengthy and complicated saga, the board of OpenAI, for unknown reasons, fired its CEO Sam Altman at 3 PM on a Friday afternoon. There was an immediate backlash because Mr. Altman is a relatively well-respected figure in his field. The backlash intensified because OpenAI is essentially a vassal of Microsoft at this point, and OpenAI didn’t bother to inform their overlords of what was going on, which meant the Great Eye of Microsoft suddenly turned upon OpenAI’s board in wrath, especially since the move might have tanked Microsoft’s stock before its quarterly report.

Meanwhile, many OpenAI leaders quit, a majority of the employees signed a letter calling for the board to resign, Microsoft immediately hired Mr. Altman and the other leaders who quit, but then the board panicked and backtracked, and finally the board quit and Mr. Altman returned. Overall, it was definitely a fascinating saga of corporate politics, and since I personally think OpenAI is like one of the evil organizations from a James Bond movie (I bet they even have an elaborate SPECTRE-style underground base somewhere), it was enjoyable to watch from afar.

But this is the blog of a pulp fiction writer, not a technology, business, or AI blog. Why write about this?

Because there is a lesson for fiction writers in this. The board of OpenAI is not stupid. They’re all intelligent men and women who are leaders in their fields. And yet whatever their goals were in firing Mr. Altman, it is readily apparent that 1.) those goals were not achieved, and 2.) the results were in fact the opposite of what they had hoped to accomplish, since they quit and Mr. Altman remained as CEO of Open AI.

And that provides a good lesson for writers of fiction.

How can you have characters make believable mistakes without breaking the suspension of disbelief?

Because when a character does something stupid solely to advance the plot, it is annoying, isn’t it? Like, the intelligent hero who suddenly becomes dumb as a brick, or the cunning villain who suddenly loses 50 IQ points. Probably the most commonly cited example is the heroine who goes into the basement with just a candle to reset the circuit breakers when she knows a serial killer or a vampire or the Terminator or something is after her. Or when the hero’s plan only works if the villain suddenly becomes much less clever.

That’s annoying in fiction because it breaks the verisimilitude.

Nevertheless, in Real Life intelligent people do dumb things all the time. Like, constantly! Examples are abundant! No doubt you can think of several dozen off the top of your head without even trying. Just as it would break verisimilitude in your fiction to have your characters be consistently idiotic, it would be just as strange to have them be infallible high-functioning geniuses.

So let’s have some tips and tricks on how to have your characters believably make bad decisions.

1.) Emotional pressure.

A key reason for many bad decisions is emotional pressure, because for most people emotions almost always trumps logic.

The most obvious example of this is a high-powered professional who has an affair with someone in his or her office, only to end up resigning in disgrace when it comes out. Once again, examples abound, and you can probably think of numerous cases from the last few years, whether national or local political figures or people you know personally who work in your organization. We can think of more positive examples – a man could have pity on a homeless man and give him his lunch, even though this means he might perform badly at an important work task this afternoon. Or a woman might be trying to save money, only for love to override her better judgment and convince her to buy a gift for her grandchild.

In fiction, you can use this is many, many ways. Both love and hatred are powerful motivators, and so are envy and resentment. A character could take dangerous risks to help someone he or she loves. Or a character could be so gripped by envy that he or she tries to sabotage a rival in a way that is self-defeating.

In romance novels, characters make decisions from emotional pressure all the time. It’s one of the staples of the genre.

A good example is the scene at the end of THE LORD OF THE RINGS when Saruman tries to stab Frodo and fails. This is, objectively, a stupid decision. Even if Saruman kills Frodo, it won’t improve his position, and if he succeeded in killing Frodo, Saruman would be immediately killed by the enraged hobbits. In fact, Frodo at that point is the only hobbit who doesn’t want to kill Saruman for his crimes, so killing Frodo would have been quite possibly the worst decision Saruman could make at that moment in time. But it makes sense in the context of the story and Saruman’s character because at this point Saruman has been devoured by hatred and resentment and cares mostly about screwing with the hobbits who (as his twisted mind sees it) robbed him of the chance to become the new Dark Lord of Middle-earth.

So long as the character’s emotional reality makes sense to the reader, decisions they make in the context of that emotionality reality, even objectively bad ones, will not seem like dumb decisions to advance the plot.

2.) Acting on bad information.

In computer science, there is a principle called GIGO – Garbage In, Garbage Out. The idea is that if you enter bad information into a computer program, the program is only going to generate bad results.

This is also very true of human decision-making. To make good decisions, you need to have good information.

For fiction writers, this means if you want to have an intelligent character make a bad decision, you can have them act on bad information.

For example, a group of fantasy heroes could be on a quest to find a magical sword that will slay the dragon terrorizing the kingdom. According to the kingdom’s wizards, the sword is in a ruined castle in the wilderness. Except the sword isn’t actually there – the castle is controlled by an evil sorcerer who magically enthralls anyone who enters it. The heroes have made a bad decision by going to the ruined castle, since they’ve gained a new enemy in the form of the evil sorcerer. But they thought they were making a good decision, but it turns out they were acting on bad information.

You can easily use this technique in non-fantasy genres as well. A detective could be misled by a witness, and waste time going down dead-ends in his investigation until he realizes the truth. In a thriller novel, the hero could realize that the informants have deliberately been feeding his agency bad information about potential threats.

Acting on bad information is also a common technique in romance novels. Usually, romance novels have a plot twist where the heroine can’t get together with her love interest for some reason, and often it’s because the heroine and the love interest misinterpret each others’ motives. PRIDE & PREDUJICE is maybe one of the oldest examples of this, since Mr. Darcy and Elizabeth Bennett immediately attribute wrong motivations to each other and it takes most of the book to work through it.

3.) Unforeseen consequences.

This is a big one, and one of the major reasons that very smart people make decisions that turn out to be bad ones. Like the example of the OpenAI corporate intrigue I mentioned above. The OpenAI board members didn’t set out to get themselves booted from the company while strengthening Sam Altman’s and Microsoft’s grip, but that’s exactly what happened.

My favorite historical example of unintended consequences is Prohibition. The US didn’t just randomly decide to wake up and vote to ban alcohol one day in 1920. The Prohibition movement in the US dated back to the early 1870s, which meant nearly fifty years of work, public relations, persuasion, and changing local laws went into what should have been Prohibition’s crowning triumph, the passing of the Eighteenth Amendment in 1920.

Except it all backfired, didn’t it? The temperance movement wanted to end alcohol consumption in the United States. What they got instead was an explosion of organized crime, increased disrespect for public authority, the loss of jobs and government tax revenue coming into the Great Depression, and public opinion turning against prohibition. The explosion of organized crime was especially ironic since many temperance advocates believed, sincerely and firmly, that the majority of all crime was caused by alcohol consumption, and that most of society’s evils could be traced to the consumption of strong liquor. Some local communities actually sold their jails after Prohibition passed, believing that crime would soon drop to near-zero.

Alas, the causes of societal evil remain multitudinous and cannot be solely hung on alcohol, and the Twenty-First Amendment passed in 1933, which was the end of national Prohibition in the United States. Prohibition’s legacy remains in the minimum drinking age (another law often ignored), restrictions on the time of alcohol sales, and “dry” counties where you can’t legally purchase or consume alcohol, but the concept of national Prohibition in the United States is obsolete.

The temperance advocates didn’t accurately foresee the consequences of their triumph, and you can use this same principle in writing fiction. A character could achieve what they set out to do, only to find that unanticipated consequences of their success are more severe than the original problem. In a thriller novel, the heroes could take out the leader of the bad guys, only for the leader’s more competent and dangerous lieutenant to take over. In a romance novel, the heroine could win a lawsuit or a big business deal, only to discover that this damages her love interest’s family business. In a detective novel, the protagonist could finally track down the key witness to the murder, only for the murderer’s attention to be drawn to that witness.

4.) Victory disease.

In military history there is a concept called “victory disease.” It happens when an army or a commander has won so many times that they have become overconfident and lazy and start making avoidable mistakes. Sooner or later they run into a more serious opponent, and an army subject to victory disease will make errors that a less complacent opponent will not. You sometimes see this is professional athletics as well – a superstar athlete or winning team gets overconfident, stops training as hard or gets complacent, and then gets their clock cleaned by a hungrier opponent.

So “victory disease” is a combination of overconfidence and complacency, and you can definitely make use of it to have a character make an understandable bad decision.

In fictions, villains tend to be more prone to victory disease than protagonists. Nevertheless, having a protagonist with victory disease can force them into internal conflicts and character growth.

A good example of victory disease in a protagonist is Batman/Bruce Wayne in THE DARK KNIGHT RISES film. When Batman comes out of retirement to fight Bane’s organization, he’s so used to winning against criminals and outwitting the police that he doesn’t take Bane seriously enough, despite Alfred’s warnings. This bites Batman hard when he confronts Bane for the first time, and he’s forced to undergo character development to get ready to save Gotham City from Bane.

You can apply a similar plot arc to your characters – an overconfident character makes a serious mistake and has to recover from it, learning and undergoing character growth in the process.

5.) Fields of expertise.

There’s a certain kind of public intellectual (they usually have “Ph.D.” in all their social media handles) that likes to pronounce upon the issues of the day. They will often say things like “as a scientist I think” or “as an academic I think” when commenting upon various issues.

What’s amusing is that their pronouncements are often wrong or wildly impractical because they are straying out of their fields of expertise. Like, they might have a Ph.D. in cellular biology or something and be a world-renowned expert in that field, but knowledge and expertise in one field does not necessarily translate to competence in another. This can result in basic errors that could otherwise be avoided.

Academics run into this a lot.

Two examples from Real Life might suffice. A reporter was covering some protests, and was alarmed to discover rubber bullets lying on the ground, and posted a picture of them to social media. He was widely mocked because the objects in question were not rubber bullets, but earplugs. A minor celebrity went to a city and was horrified to see racist graffiti on the sidewalk and complained about to social media, only for (many) commentors to point out that the symbols were not graffiti at all but markings from utility workers indicating where electrical and gas lines ran underneath the sidewalk.

This makes for a very believable way for your characters to make bad decisions – force them to make decisions in an area where they don’t really know what they’re doing.

6.) Conclusion

To maintain verisimilitude in fiction, you need to walk a fine line – if your characters suddenly become stupid to advance the plot, that will annoy the readers, but neither can your characters be infallible reasoning machines. Hopefully these tips and tricks will help your characters make mistakes in a believable way.

-JM