Chapter 18 — Key Takeaways

One page. 🚪 Threshold chapter.

🚪 The Threshold: the dose makes the poison — both halves

Sola dosis facit venenum. Paracelsus, sixteenth century.

So is sugar bad? is not a question. Neither is is fat bad?, are carbs bad?, is coffee good? They are category errors, which is why decades of evidence haven't settled them.

What replaces them — five slots:

At what dose · in what form · over what period · in whom · compared to what?

⚠️ It's uncomfortable in both directions — no villain for the enthusiast, no all-clear for the sceptic. That's how you know it's a threshold and not a slogan.

🧩 Part IV is this one concept applied five times: Ch 19 (seed oils — a dose question in a purity costume) · Ch 20 (residues — a dose question argued as a presence question) · Ch 21 (fasting — a dose question about time) · Ch 22 (UPF).

The Four Claims, argued as one

⚠️ The move to watch, in both directions: argue for Claim 1 and treat Claim 4 as established. Or refute Claim 4 and treat Claims 1–3 as disposed of.

The Chemistry, briefly

Sucrose = glucose + fructose, bonded, cleaved in the small intestine within minutes. HFCS-55 = ~55% fructose / ~42% glucose, free. ⚠️ By the liver, near-indistinguishable. Agave = ⚠️ often 70–90% fructose — usually the highest-fructose product in the shop.

§18.4 — The reframe that matters most

💡 ⚠️ This is why "is fructose harmful?" has no answer, and why both camps are frequently right about different doses. One is describing high-dose overfeeding studies; the other is describing intakes reachable from fruit. Neither is lying. They have not agreed on the dose.

Two honest qualifications on DNL: it's a modest contributor to whole-body fat balance at ordinary intakes (most body fat was eaten as fat — Ch 6 §6.7); and ⚠️ most high-dose fructose studies are also overfeeding studies, so they can't separate Claim 2 from Claim 3. At matched calories the differences shrink substantially — whether to zero is exactly what's unresolved.

Verdicts

⚠️ §18.7 — Take this over any of the arguments

People do not compensate for liquid calories. Eat 200 kcal solid → you eat somewhat less later. Drink 200 kcal → compensation is close to zero.

Why: faster gastric emptying · less distension · weaker satiety hormone response (Ch 3 §3.6) · no chewing · consumed alongside other activity rather than as an eating event.

⚠️ Theo drank two sodas a day and did not experience himself as having eaten anything. Nobody fails to notice a doughnut.

Note what the ✅ attaches to: a delivery format, not a molecule.

§18.6 — Where the alcohol analogy holds and breaks

Holds: hepatic metabolism · steatosis at dose · liquid, poorly compensated, heavily marketed · NAFLD is a real and growing problem.

Breaks: ⚠️ ethanol produces acetaldehyde, a direct cellular toxin — fructose has no counterpart · the intestine buffers fructose and does not buffer ethanol, so the low-dose curves differ in shape · ethanol is psychoactive and causally implicated in injury and death · ⚠️ Chapter 12 had ALDH2 as a causal instrument; sugar has no equivalent.

Rejecting the analogy is not the same as rejecting the concern.

⚠️ §18.12 — The hardest lesson in the chapter

Documented (Kearns, Glantz & Schmidt, JAMA Internal Medicine, 2016, from recovered correspondence): the Sugar Research Foundation funded a 1967 NEJM review that downplayed sugar and emphasized fat, with undisclosed funding and sponsor input. ✅ Real.

Over-extended: that it single-handedly caused the low-fat era. ⚠️ Four qualifications: non-disclosure was normal in 1967 · Keys's work was independent · Yudkin's suppression story is embellished · the low-fat era had many parents (Ch 9 §9.7).

⚠️ This looks exactly like the suppressed-truth shape Chapter 17 taught you to dismiss — and it's true.

The test: a suppressed-truth claim is unfalsifiable and cites nothing. A documented episode names the year, the journal, the authors, the sponsor, and the archive. Carry this into Ch 19 and Ch 20.

§18.13 — Sweeteners, and a structure worth learning

⚠️ 2023: IARC classified aspartame Group 2B ("possibly carcinogenic"); JECFA reaffirmed the ADI of 40 mg/kg/day. Same week, reported as contradictory. They answered different questions:

Group 2B also contains aloe vera extract and pickled vegetables. ⚠️ This is Ch 17 §17.9's BPA structure exactly — learn the shape, it's worth more than the verdict.

On weight: WHO 2023 issued a conditional recommendation against NSS for weight control, on low-certainty, largely observational evidence with obvious reverse causation. RCTs substituting NSS for sugar generally show modest benefit. ⚠️ The trials and the cohorts disagree and the WHO weighted the cohorts.

Verdict 🟡 — and the comparator is the whole question: versus soda, likely an improvement. Versus water, no case.

What To Actually Do

1. ⚠️ Deal with the liquid first. 2. Find your top three sources before changing anything. 3. Look at what doesn't taste sweet. 4. ⚠️ Don't switch syrups — cane/agave/honey is not a change. 5. Whole fruit isn't the problem; ⚠️ juice is closer to soda than to fruit. 6. Use a target, not a prohibition — WHO <10% of energy (conditionally <5%); AHA ≈25 g women, ≈36 g men. 7. ⚠️ Check the denominator — what replaced it?

🧾 The Economics

Ortiz-Lindqvist household: −54% added sugar, −$470/year. The only line that cost more was plain yoghurt + frozen berries (+$1.20/week).

Fifth chapter running where the evidenced change is the cheaper one.

One Thing to Remember

Theo's desserts were 8% of his added sugar. Two sodas were 50%.

⚠️ The part he'd have given up first was the part that didn't matter — and half of it was something he never experienced as eating at all.

Find your sources before you change anything.