Logical fallacies abound. They’re harmful. You have a new tool to help you uncover them.
Talking Big Ideas.
“Fix reason firmly in her seat,
and call to her tribunal every fact, every opinion.”
Bob Menendez is having a tough week.
Most of his fellow Democratic Senators have called for him to resign in the wake of a federal indictment accusing Menendez of secretly aiding a foreign government, including conspiracy to commit bribery, extortion, and fraud. The story is wild, with arms sales to Egypt, secret gold bars, the Halal meat industry, a Mercedes convertible, lots of deleted text messages and more. In response to the blistering 39-page indictment, the Senator said:
It is not lost on me how quickly some are rushing to judge a Latino and push him out of his seat.
This is a classic example of a logical fallacy called an Ad Hominem, which is Latin for “against the person.” Whenever someone attacks the person rather than addressing the merits of the argument being made, it’s an ad hominem fallacy.
This week we also had a presidential debate, which are goldmines for logical fallacies. Chris Christie came out slinging an ad hominem at Trump (“Donald Duck”) and the line of the night was this ad hominem: “every time I hear you, I feel a little bit dumber.”
During just one 2020 debate, Trump and President Biden committed 15 different logical fallacies together. Importantly, their rhetoric was mild compared to the often-cherished era of our founding fathers. The election of 1800 sounded like this:
[If Thomas Jefferson wins] murder, robbery, rape, adultery, and incest will be openly taught and practiced, the air will be rent with the cries of the distressed, the soil will be soaked with blood, and the nation black with crimes.
We all understand that politics is a dirty business. Most of us do our best to ignore the mudslinging, and our guard is up when we hear politicians speak. It can be harder to spot fallacies that sneak into our minds in more insidious ways.
For example, tobacco companies used to feature doctors endorsing their cigarettes. If doctors smoke, then surely smoking must be healthy, right?
This is an Appeal to Authority fallacy, which happens whenever we exploit the public’s trust in high-status individuals while ignoring contrary evidence. As the scientist Richard Dawkins likes to say, “An argument from authority is no argument at all.” Good arguments use logic and evidence.
Millions of people died after being misled about the dangers of smoking. Fallacies are bad because they bury the truth and erode critical thought. They can manipulate, polarize, and deceive – sometimes with deadly results.
Clear and rational thinking is essential to moral as well as material progress.
Thankfully, many advertisements use humor to help us spot fallacies. Direct TV’s commercial, Don’t Wake Up in a Roadside Ditch, is a funny and fantastic illustration of the Slippery Slope fallacy:
A relatively small first step will inevitably lead to a chain of negative consequences, without sufficient evidence for the causal relationship. In this case, if you don’t buy Direct TV, you’ll end up in a roadside ditch.
How about this classic ad from Old Spice:
The commercial weaves together an appeal to authority (the handsome actor Isaiah Mustafa) with a slippery slope – get your man to use Old Spice and a bunch of amazing things will happen.
What about all the social media and news sites we encounter? How can we get better at detecting fallacies in them?
But now we have amazing new fallacy-detection tools: ChatGPT, Google Bard, and the other AI LLMs (Chat). Simply copy and paste text from any media story and they will analyze for fallacies. I like to post the same questions to both ChatGPT and Google Bard, and then verify the answers they give me.
Consider a simple prompt like this:
Chat, analyze this news story, and tell me if it commits any logical fallacies. If so, clarify them with specific examples.
I just grabbed a news story off the front page of NBC News about Taylor Swift, pasted it into Chat, and included the prompt above. Chat replied:
The news story relies on anonymous sources and speculation about Swift’s personal life, which could be considered a breach of ethical journalism standards, but it doesn’t seem to overtly commit logical fallacies.
Chat then gave me three ways the article may mislead readers:
- Anonymous Sources
With each bullet Chat included specific examples from the piece and explained how they can be deceiving. You can read my question and Chat’s full answer here.
More importantly, two weeks ago the L.A. Times covered a debate on the sexual revolution hosted by The Free Press and the Foundation for Individual Rights and Expression. As I read the L.A. Times story I was taken aback by its snarky tone. So I copied the text of the story into Chat along with the prompt above.
Chat’s full response is here. You can see how it analyzed the piece and discovered eight logical fallacies:
- Ad Hominem Attacks
- Straw Man Fallacy
- Begging the Question
- Tu Quoque (You too) Fallacy
- Appeal to Ridicule
- Association Fallacy
- Red Herring
The L.A. Times is among the most widely read papers in the country. It’s won more than 40 Pulitzer Prizes! If the L.A. Times is willing to publish this article filled with fallacies, what about everything else we are consuming?
Within the next two weeks, ChatGPT will be able to see, hear, and speak. You can simply share audio clips as well as pictures, graphs, charts, ads, books, images, social media posts, and other data to quickly get a detailed breakdown on any inaccuracies.
You can also have Chat check your own writings, speeches, and thoughts. Before I publish anything on this Substack now I have Chat review it for edits and accuracy. And I’ll ask Chat to review various ideas I have, and conversations I’ve held.
Chat is an incredible resource for ensuring your ideas are sound. As the physicist Richard Feynman said in his iconic 1974 Commencement Address to Caltech:
The first principle is that you must not fool yourself — and you are the easiest person to fool.
I encourage you to use Chat to find fallacies in the wild . . . and make sure you honor Feynman’s first principle.