Case Study 2 (Deep Dive): When an Analogy Breaks at Scale

A deep dive into the chapter's most important and most dangerous idea: that every analogy breaks, and an unflagged break can mislead millions. The examples here are real and widely discussed; we keep to what's verifiable and avoid inventing detail.

Why this case

Case Study 1 showed an analogy doing its job. This one shows the other half of §28.3—the half that separates a science communicator from someone who just likes vivid comparisons. The thesis: a good analogy that isn't flagged where it breaks doesn't just fail; it actively installs a wrong idea in the reader's head, and the better the analogy, the deeper the wrong idea goes. We'll look at three analogies that are genuinely useful, genuinely widespread, and genuinely break in ways that matter—and at what the writer should do about each.

The pattern to watch for is the one from the chapter's threshold concept: an analogy is a loan against the reader's existing understanding, and the reader will spend that loan further than you intended. The break is the interest. Whoever deployed the analogy owes the disclosure.

Break 1: "A computer virus is like a biological virus"

This is one of the most successful analogies in all of technology—so successful that we no longer notice it's an analogy. We say a program "infects" a machine, "spreads," is "contagious," that you need to "quarantine" it and run "antivirus." The source (biological infection) maps beautifully onto the target (self-replicating malicious code): something small gets into a host, copies itself, spreads to other hosts, and causes harm. For explaining what malware does and why it spreads, the analogy is excellent and worth keeping.

Where it breaks, and whether it matters. Run the two-step check (§28.3).

• A biological virus spreads through involuntary, physical contact—you can't choose not to breathe near someone. Much malware spreads through a human choice: clicking a link, opening an attachment, installing software. This break is load-bearing for a security-advice article, because it changes the entire prevention model. If a reader fully buys "it's like the flu," they conclude that infection is something that happens to them, like weather—when in fact the single most effective defense is a behavior they control (don't click that). An article on staying safe online that leans on the virus analogy without flagging this break has, however unintentionally, taught its readers learned helplessness.
• A biological virus mutates on its own; malware is deliberately written and modified by people. This break matters if the article is about who's attacking you and why (there's an adversary with intent, not a mindless pathogen)—and it's invisible if the article is just explaining replication.

The fix. A security writer using the virus analogy for a general audience should keep it (it's too good to discard) and add one sentence flagging the load-bearing break: "Unlike a real virus, most computer infections need you to let them in—a click, a download—which is also the good news, because it means the best protection is something you control." That single flagged break converts the analogy from one that disempowers the reader to one that empowers them, without abandoning the familiar source. Same analogy; the disclosure changes what the reader does on Monday.

Break 2: "The brain is a computer" (and "memory is storage")

This analogy organizes how a generation talks about the mind. We "process" information, "store" memories, "retrieve" them, run out of "bandwidth," "encode" and "delete." It's a useful source for a general reader because computers are familiar and the comparison conveys real things: the brain takes in information, holds it, and acts on it.

Where it breaks. The breaks here are deep, and several are load-bearing for almost any article about how memory actually works.

• A computer stores a file and returns it unchanged; human memory is reconstructive—each time you recall something, you partly rebuild it, and it can change in the process. This is the load-bearing break for any piece about eyewitness testimony, false memories, or studying, because the whole point of those topics is that human memory is not a faithful playback. A reader who takes "memory is storage" literally will misunderstand the single most important fact about it.
• A computer's storage and processing are separate; in the brain they're deeply intertwined. This matters for articles about learning, less so for a casual aside.

The fix. This is a case where, if the break is your point, you should switch sources or foreground the break rather than paper over it. An article about false memories might open by using the reader's computer assumption against them: "You probably think of memory like a video recording—press play and watch what happened. It isn't. Every time you remember something, your brain rebuilds it from fragments, and like any rebuild, it can come back slightly wrong—or completely wrong." Here the writer names the broken analogy explicitly and uses the reader's prior (wrong) model as the hook (§28.4)—a sophisticated move: the curiosity gap is the gap between the familiar analogy and the surprising truth. The break becomes the story.

Break 3: "DNA is the blueprint of life"

A staple of popular biology, and a good first handhold: DNA contains instructions that shape an organism, the way a blueprint contains instructions for a building. For "DNA carries hereditary information," it works.

Where it breaks. A blueprint is a static, complete, one-to-one plan—every line on the blueprint corresponds to a feature of the building, and the building is read off the blueprint once. DNA is not like that in several ways that matter: the same gene can do different things in different contexts; the environment switches genes on and off (the whole field of epigenetics); there's no tidy one-to-one map from "gene" to "trait"; and the cell is constantly reading and re-reading the DNA, not building from it once. For an article about "a gene for X" (intelligence, height, a disease), the blueprint analogy is actively misleading, because it implies exactly the deterministic one-gene-one-trait picture that modern genetics spends most of its time correcting.

The fix. Either flag the break hard or change the source. A common better source is a recipe: a recipe gives instructions, but the result depends on the cook, the oven, the ingredients on hand—so the same recipe yields different cakes, the way the same genes yield different outcomes in different environments. The recipe analogy maps the context-dependence that the blueprint analogy hides. A writer choosing between them should pick based on the §28.3 question: does my point depend on the part the source gets wrong? If the article is about heredity in general, the blueprint is a fine first handhold (flag it lightly). If the article is about why "a gene for intelligence" is a misleading phrase, the blueprint's break sits exactly on the point—so reach for the recipe instead.

The throughline

Put the three together and the lesson is unmistakable, and it's the chapter's central skill stated three times:

In every case the analogy is good—worth using, familiar, illuminating. And in every case, used without attention to its break, it installs a false belief that the reader will carry confidently because you gave it to them with your authority attached. That's why §28.3 calls finding the break the most important skill in the chapter, not the analogy itself. Anyone can reach for a vivid comparison. The communicator's discipline is knowing exactly where it stops being true, deciding whether that edge sits on the point, and—when it does—telling the reader before they walk off it.

There's an ethical edge here too, which Chapter 38 will sharpen: a deliberately unflagged break is how clear writing sells a wrong idea. The same persuasive power that makes the gym analogy land can make "it's just like the flu, nothing you can do" land—and the second is false. Clarity is a responsibility precisely because it works. The honest communicator borrows the reader's understanding with an analogy and discloses the cost. The dishonest one borrows and never mentions the bill.

Back to: Chapter 28 · Case Study 1 · Exercises · Key Takeaways