Chapter 17 — Key Takeaways
One page. Opens Part IV.
The Eight Shapes — and their tells
⚠️ Every tell is structural. Not one requires you to know any nutrition.
| Shape | Tell |
|---|---|
| Single cause | No dose, no comparator. "Is X bad?" with no "compared to what?" |
| Purity and contamination | Moral vocabulary — clean, pure, toxic, poison, chemical |
| Restriction as virtue | The difficulty is offered as evidence. "It's not easy, but…" |
| Suppressed truth | Unfalsifiable — absence of evidence becomes proof of suppression |
| Lost golden age | The past is unspecified. Which ancestors? Where? When? |
| Hidden simple fix | Mechanism named, never dosed |
| Dial described as a switch | Binary language, continuous variable — mode, reset, unlocked |
| Bundled intervention | ⚠️ **Count the changes. More than three and nobody can learn anything — |
| including the seller.** |
⚠️ The Central Insight: over-extension, not fabrication
Twenty-four myths were already retired in Chapters 1–16. Only six were ❌. The rest were 🟠 or 🟡 — a real finding stretched past what it could carry.
So the productive question is almost never "is that nonsense?" It's "what's the narrower true version?"
Protein synthesis does plateau. Fat oxidation does vary with intensity. Saturated fat does raise LDL-C. Refined carbohydrate does make overeating easy. Every one became a myth by being stretched.
A reader who learns that question will out-perform a reader who memorizes fifty verdicts.
Verdicts
| Claim | Verdict |
|---|---|
| Detoxes and cleanses remove toxins | ❌ (Ch 3) — ⚠️ the toxins are never named, so the claim cannot fail |
| Superfoods have special properties | ❌ Not a scientific or regulatory category. USDA withdrew the ORAC database in 2012 |
| The alkaline diet changes blood pH | ❌ Blood pH is held at 7.35–7.45. ⚠️ Urine pH changing proves your kidneys are working |
| Celery juice heals chronic illness | ❌ Rung 3 only; origin explicitly non-empirical. ⚠️ Targets people who may delay real treatment |
| Supplements "boost" your immune system | ❌ Describes autoimmunity, not health. 🟢 Correcting a real deficiency does help |
| Negative-calorie foods | ❌ Celery: ~6 kcal in, ~0.5 kcal to digest. Net +5.5 |
| The blood type diet | ❌ Tested directly (PLOS ONE 2014) — benefit was unrelated to blood type |
| Microwaving destroys nutrients | 🟠 Backwards — shorter time, less water, often retains more than boiling |
| Non-stick pans poison you | 🟡 ⚠️ PFAS genuinely matter; intact pans at normal temperature mostly don't. Real hazard = overheating an empty pan (>~260°C) |
| Plastics leach hormone disruptors | 🟡 A live regulatory disagreement — EFSA lowered its TDI in 2023; FDA maintains safety. ⚠️ "BPA-free" often means BPS/BPF, less studied |
| Cooked tomatoes beat raw for lycopene | ✅ Well supported — cell walls broken down, better with fat |
| Ginger helps nausea | 🟢 Especially pregnancy-related; in several clinical guidelines |
| Rinse rice / cook in excess water | 🟢 Starch and inorganic arsenic — ⚠️ matters most for infants and rice-staple diets |
| Chicken soup helps a cold | 🟢 Modestly. Doesn't shorten it. "Feels better" is a legitimate outcome |
⚠️ The Hardest Section: §17.9
Detoxes, superfoods, alkaline and blood-type diets have no true version. The contamination cluster does.
| The false version | The true version |
|---|---|
| "Microwaves destroy your food" | Cooking method affects nutrients — and microwaving usually wins |
| "Your pan is poisoning you" | PFAS are a genuine, under-addressed environmental contaminant |
| "Plastic is full of hormone disruptors" | BPA is genuinely unresolved and two regulators disagree |
⚠️ Dismissing the false version can make you dismiss the true one. This is the failure mode a sceptical reader is most prone to, and it's why 🟡 exists on the scale.
Cheap responses under uncertainty: don't heat food in plastic · replace scratched containers · don't preheat empty non-stick pans · ventilate. These cost nothing, which is what makes them reasonable without resolving the science.
Why the true ones are true
§17.10's four supported claims share a shape: small, specific, dosed, checkable claims about a mechanism. None cures anything. None requires a category of person. None has a suppressed-truth story.
Modest claims survive scrutiny at a far higher rate than grand ones. Close to a general rule.
Why Myths Stick (§17.11)
They resolve uncertainty · offer a visible confirmation (urine strips, ketone meters, beta-alanine tingling) · provide agency · confer identity and community · are morally legible · produce genuine reported results through bundling — ⚠️ nobody is lying about how they feel · and are nearly unfalsifiable.
How Myths Die (§17.12)
Trans fats → regulation. Ephedra → deaths, then a ban. Protein combining → retracted by its own author, over forty years. Eight glasses → debunked definitively in 2002, still with us.
⚠️ Myths die from regulation and counted harm. They almost never die from being refuted.
When Someone You Love Believes One (§17.13)
Triage first:
| ⚠️ Intervene | Probably let it go |
|---|---|
| Delaying or replacing effective medical treatment | Celery juice |
| Real harm profile — high-dose supplements, colon cleansing, extreme restriction | A January cleanse |
| Restriction in a child, a pregnancy, or an eating-disorder history | Expensive but affordable |
| Significant money from someone who hasn't got it |
Then: compare, don't debunk (comparison doesn't require them to have been wrong) · ask about the mechanism, not the conclusion · decompose the bundle — "what else changed?" · concede the true part first, and mean it · ⚠️ never make it about intelligence — Theo has a folder, two highlighters, and eight of ten wrong · and expect to fail, while leaving a marker.
The Ten-Minute Check (§17.14)
0–1 Restate it so it could be wrong. ⚠️ If you can't, that's the finding. 1–3 Run the eight tells — not to decide truth, to decide how much scrutiny it's worth. 3–5 ⚠️ Rung supplied vs rung required — the most diagnostic number available to a non-specialist. 5–7 Denominator and direction. Compared to what? Healthy-user bias? 7–9 Source: what happens to their income if you decline? Do they ever say "I don't know"? 9–10 Assign a verdict — including ⚗️ and 🟡, which are correct more often than anyone finds comfortable.
⚠️ It's a good filter and a weak detector. Ten minutes can reject; it cannot establish truth.
Two Categories Worth Naming
Stale guidance (Case Study 1) — correct when issued, superseded since, never retracted to the person holding it. Theo's eggs advice came from a doctor. ⚠️ There is no mechanism in the health system for telling you that something you were told in 2009 is no longer believed.
Ambient belief — no attributable source. ⚠️ Source-evaluation doesn't work on "everyone," which is why these were 4/4 wrong on Theo's sheet.
One Thing to Remember
Four of the five beliefs Theo rated maximum confidence were wrong. The only one he was right about was the only one he doubted.
⚠️ His confidence tracked repetition, not evidence — and repetition is exactly what a myth is good at.
That's a fact about how belief is formed, not about nutrition. It will still apply to claims that don't exist yet.