Chapter 3 — Digestion: From Mouth to Bloodstream — How Your Body Breaks Down, Absorbs, and Uses Food

The Hook: The boy who wouldn't eat dinner

Erik Lindqvist makes dinner four nights a week, because his wife Camila works three twelve-hour night shifts and somebody has to.

On the Tuesday he described to me, he made rice, black beans, roasted sweet potato, and chicken. Alma — four years old, peanut-allergic, and a connoisseur of exactly nine acceptable foods — ate the rice and negotiated over the sweet potato. Nico, who is nine, ate about a third of his plate and then put his fork down and said his stomach hurt.

Again.

It had been happening for maybe five months. Not every day. Not severe enough for the emergency room. Just often enough that Erik had started watching his son eat the way you watch a car making a noise.

A neighbor — kind, well-meaning, absolutely certain — told them it was gluten. She'd had the same thing. Cut it out and everything changed.

So they cut it out. Different bread, different pasta, checking labels, an extra forty dollars a month they didn't have. And it seemed to help, a bit, some weeks. Which made them more certain, which made them stricter.

By the time the family reached my colleague Yolanda four months later, Nico had been off gluten long enough that the test for celiac disease would no longer work on him — because that test requires gluten in the diet to be valid. They had eliminated the thing before testing for it, which is the single most common and most costly mistake in this entire area, and nobody had told them.

We'll follow Nico properly in Chapter 28. What I want here is the thing Erik said when Yolanda started explaining.

He said: "I don't actually know what happens to food. Like — at all. It goes in, and then I guess something happens, and that's it."

He's a high-school teacher. He is not an uneducated man. And he was describing, honestly, the mental model that most adults have of the thirty-foot tube that determines what their food is actually worth to them.

You cannot evaluate a claim about food without knowing roughly what happens to it. That's the case for this chapter, and it's more practical than it sounds.

Almost every "detox" claim dies on contact with what a liver actually does. Food combining rules dissolve once you know where enzymes are secreted. The "you can only absorb 30 grams of protein per meal" rule that has shaped a generation of gym eating turns out to confuse two different processes. The reason fiber does anything at all is a story about the large intestine. The reason lactose intolerance is the global norm rather than a disorder is a story about one enzyme on the surface of one type of cell. The reason bariatric surgery works far better than its calorie arithmetic predicts is a story about hormones you've never heard of.

All of that is in this chapter. So let's follow a sandwich.

🏃 Fast Track: §3.1 (the tube), §3.4 (the small intestine — where nearly everything happens), and §3.7 (gut hormones) carry most of the value. §3.2 and §3.3 are short.

🔬 Deep Dive: §3.5 (first-pass metabolism and why fat takes a different route) and §3.8 (what goes wrong) are the sections students and clinicians should sit with. §3.6 is the foundation for Chapters 11, 22, and 27.

3.1 You are a donut

Start with the strangest true thing about your digestive system.

The inside of your gut is outside your body.

Topologically, a human being is a torus — a donut. There's a tube running from your mouth to your anus, and everything in that tube is in continuous contact with the outside world. Food in your stomach has not entered you. It's in a chamber that happens to be surrounded by you.

Nothing you eat is in your body until it crosses the gut wall into your blood or lymph.

This is not a philosophical point. It's the organizing fact of the whole chapter, and it immediately reframes three things:

1. Eating and absorbing are different events. You can eat something and absorb none of it. This is exactly what happens with most fiber, with the sugar alcohols in sugar-free products (which is why they cause digestive complaints — they arrive undigested at a fermentation chamber), and with nutrients you consumed in a form your body can't take up.

2. Your gut is a border, and borders need security. Something like 70% of your immune tissue sits along the digestive tract, because that's where the outside world is closest. This is why food allergy is an immune event, why the gut is central to inflammatory disease, and why the microbiome — which lives on the far side of that border — matters.

3. "Cleaning out your colon" is a category error. The colon is not a storage vessel of accumulated sludge. It is a functioning organ with a continuous throughput, doing active work.

📊 Diagram (described). Picture a garden hose about nine metres long, running from top to bottom through the middle of a person, wider in some places than others. At the top, a short wide section (mouth and esophagus) that just moves things along. Then a J-shaped bulge (the stomach) that holds and churns. Then, below it, the long narrow section that does the actual work — the small intestine, roughly six to seven metres, coiled tightly, its inner surface not smooth but thrown into folds, and those folds covered in finger-like projections (villi), and each of those covered in smaller projections (microvilli), like a shag carpet made of shag carpet. Then a shorter, wider section (the large intestine, about 1.5 metres) that mostly reclaims water and hosts an enormous microbial population.

Now the key visual: run a line down the centre of the hose and shade everything inside it as "outside the body." The nutrients only count when they cross the hose wall. Everything before that crossing is preparation.

🔍 Why this works. Why bother with the villi-on-folds-on-villi architecture? Surface area. Absorption happens across a membrane, so absorptive capacity is proportional to area. A smooth tube of the small intestine's dimensions would offer perhaps a third of a square metre. Folding it, covering the folds in villi, and covering the villi in microvilli multiplies that enormously — older textbooks compare the result to a tennis court, though more careful recent estimates put it closer to 30–40 square metres, roughly the floor of a small apartment. Either way, the principle holds: your body solved the absorption problem with geometry. And this is why damage to the villi — as in untreated celiac disease — causes malabsorption of many nutrients at once. You haven't lost an enzyme. You've lost carpet.

3.2 The mouth: more than chewing

Erik's chicken and rice enters the mouth. Three things happen, only one of them obvious.

Mechanical breakdown. Chewing increases surface area for enzymes. This matters more than it sounds: people who chew less, or who have poor dentition, extract measurably less from some foods. It's one reason Chapter 25 takes teeth seriously as a nutrition issue in older adults.

Chemical breakdown begins. Saliva contains amylase, an enzyme that starts breaking starch into smaller sugars. (Hold a piece of bread in your mouth for ninety seconds without swallowing and it turns sweet. That's amylase, working in real time, and it's the cheapest biochemistry demonstration available.) There's also a small amount of lingual lipase for fat.

The cephalic phase. This is the interesting one. Before food arrives, the sight, smell, and even thought of it triggers preparation — saliva, stomach acid, pancreatic enzymes, and a small early release of insulin. Your body starts digesting before there's anything to digest.

The cephalic phase is why food that smells good is genuinely easier to digest, why eating while distracted changes the experience, and — a thread we'll pick up in Chapter 33 — why the reward system and the digestive system are not separate departments.

3.3 The stomach: what it actually does

Ask most people what the stomach does and they'll say "digests food." It does some of that. It is not the main event, and understanding its actual job resolves several myths.

What the stomach does:

What the stomach does not do: absorb much of anything. A little water, some alcohol, a few drugs. Essentially no nutrients. The stomach is a preparation chamber, not an absorption site.

Gastric emptying: the underrated variable

How fast the stomach empties into the small intestine shapes a great deal.

• Liquids empty fast. A glass of juice hits the small intestine quickly. This is a large part of why liquid sugar behaves differently from the same sugar in whole fruit — see Chapter 18.
• Fat slows emptying substantially. Fat in the duodenum triggers hormonal brakes.
• Protein and fiber slow it too, though less dramatically than fat.
• A mixed meal takes roughly two to four hours to leave the stomach, with wide individual variation.

This is the actual mechanism behind a lot of "satiety" advice. When people say protein and fiber keep you full, part of what they're describing is a stomach that stays occupied longer, plus the hormonal signals in §3.7. It's also why the drugs in §3.7 work partly by slowing this down.

The intrinsic factor detail matters clinically. People who've had stomach surgery, who have autoimmune damage to parietal cells, who take long-term acid-suppressing medication, or who are simply older and producing less acid can absorb B12 poorly even with plenty in the diet. Nutrient intake and nutrient status are not the same thing — the Chapter 13 threshold concept, arriving early.

🔄 Check your understanding. Someone tells you a food "sits in your stomach and rots, releasing toxins." Give two reasons from this section that this can't be right.

Answer

(1) pH. The stomach runs at roughly 1.5–3.5 — acidic enough to kill most bacteria. Putrefaction requires microbial action, and the stomach is a hostile environment for it. (2) Throughput. The stomach empties into the small intestine within a few hours; nothing sits there long enough to rot.

Bonus: the "toxins" are never named, which as Chapter 2 §2.7 notes is characteristic — an unnamed toxin can't be measured, so the claim can't be tested. Genuine problems with delayed emptying (gastroparesis) exist and are real, and they present as nausea, vomiting, and fullness — not as vague toxicity requiring a cleanse product.

3.4 The small intestine: where everything happens

If you remember one section of this chapter, make it this one. Essentially all nutrient absorption happens here, in about six to seven metres of tube, over roughly two to six hours.

Chyme arrives in the duodenum — the first stretch — and immediately triggers two deliveries.

Delivery one: the pancreas

The pancreas secretes a bicarbonate solution (neutralizing stomach acid, because the enzymes coming next don't work in acid) plus a full enzyme kit:

• Pancreatic amylase — starch → shorter sugar chains
• Proteases (trypsin, chymotrypsin, and others) — proteins → short peptides
• Pancreatic lipase — triglycerides → fatty acids and monoglycerides

This is the main enzymatic event. The mouth and stomach were preliminaries.

Delivery two: bile

Bile is made by the liver, concentrated and stored in the gallbladder, and released into the duodenum when fat arrives. It is not an enzyme. It's a detergent.

Fat and water don't mix, and the inside of your gut is watery, so dietary fat arrives as large droplets with very little surface exposed. Bile salts break those droplets into a fine emulsion — exactly what dish soap does to grease — massively increasing the surface area available to lipase.

This is why people without a gallbladder often struggle with large fatty meals: they still produce bile, but they've lost the reservoir that delivers a concentrated squirt on demand.

The brush border: the last step

Now the elegant part. The final digestive enzymes aren't floating in the gut — they're anchored to the surface of the intestinal cells themselves, on the microvilli that form the "brush border."

• Lactase — lactose → glucose + galactose
• Sucrase-isomaltase — sucrose → glucose + fructose
• Maltase — maltose → glucose
• Peptidases — short peptides → amino acids

Digestion finishes at the doorstep and absorption happens immediately.

And here is the whole of lactose intolerance, in one line. Lactase is a brush-border enzyme. Most humans stop producing much of it after weaning, because that's the mammalian default. If lactose isn't split, it isn't absorbed; it travels on to the colon, where bacteria ferment it, producing gas and drawing in water. Bloating, cramps, diarrhea. Not an allergy. Not damage. A missing enzyme at a specific location — and the global majority state, which we'll return to in Chapter 28.

What gets absorbed, and how

Note the fat route. Carbohydrate and protein go straight to the liver via the portal vein, so the liver gets first look at everything. Long-chain fats do not — they're packaged into chylomicrons and enter via the lymphatic system, joining the bloodstream near the heart. Fat enters general circulation before the liver processes it.

This asymmetry explains real things: why fat-soluble vitamins and some drugs behave differently from water-soluble ones, why blood triglycerides rise for hours after a fatty meal, and why alcohol — absorbed early and sent straight to the liver — is metabolically a special case (Chapter 12).

🔬 Claim → Evidence → Verdict

The claim: "Your body can only absorb 20–30 grams of protein per meal. Anything more is wasted, so you must spread protein across five or six meals."

Where it comes from: A real finding, misdescribed. Studies of muscle protein synthesis found that the MPS response to a protein dose plateaus somewhere in the region of 20–40 grams of high-quality protein in young adults — beyond that, additional protein doesn't further increase the rate of muscle building from that meal. That's a genuine result, and it's about muscle building.

What the evidence actually shows: It got translated into "absorption," which is a different process entirely. Your small intestine will absorb essentially all the protein you give it, given time — it just takes longer for a larger dose, because gastric emptying slows and transit extends. The amino acids that don't drive muscle protein synthesis aren't excreted; they're used for every other thing protein is used for (enzymes, hormones, immune proteins, gut lining turnover, connective tissue) or oxidized for energy. Trials comparing the same daily protein spread across many meals versus concentrated in fewer generally find small or no differences in body composition outcomes. Also worth noting: the plateau figures come largely from young adults and shift upward in older adults, which is Chapter 25's territory.

📉 Evidence quality: The underlying MPS studies are rung 6, small, and short, measuring a surrogate (synthesis rate) rather than an outcome (muscle gained). The extrapolation to "absorption" has no support at all.

Verdict: 🟠 Probably false as stated. Total daily protein matters far more than distribution. Spreading protein across meals is mildly, plausibly beneficial for maximizing MPS; the claim that protein above 30 g is wasted confuses absorption with utilization and has cost a lot of people a lot of Tupperware. Full treatment in Chapter 8.

🔬 Claim → Evidence → Verdict

The claim: "Don't combine protein and starch in the same meal — your body can't digest them together, and the undigested food putrefies."

Where it comes from: A dietary system dating to the early twentieth century, built on a plausible-sounding premise: protein digestion needs acid, starch digestion needs alkaline conditions, therefore they interfere.

What the evidence actually shows: The premise misunderstands where digestion happens. Salivary amylase is inactivated by stomach acid, yes — but starch digestion resumes in the small intestine via pancreatic amylase, in a bicarbonate-neutralized environment, at the same time as protein digestion by pancreatic proteases. They're designed to run simultaneously. Every whole food is already a mixture: beans, grains, milk, and meat all contain more than one macronutrient. Trials of food-combining diets have generally found no advantage over calorie-matched conventional diets.

📉 Evidence quality: Mechanistically refuted; the small trial literature is null.

Verdict: ❌ Not supported. Your small intestine digests everything at once. That's its job.

🧩 Productive struggle. Take four minutes on this before reading on.

A person has surgery that removes or bypasses a large section of their ileum — the last part of the small intestine.

Using only what's in §3.4, predict at least three specific nutritional consequences. Be as specific as you can about which nutrients and why.

What I'd expect

1. Vitamin B12 deficiency. B12 absorption is site-specific: it happens in the ileum, bound to intrinsic factor. Lose that stretch and dietary intake becomes irrelevant. These patients generally need B12 by injection or high-dose oral routes, monitored for life.
2. Fat malabsorption and fat-soluble vitamin deficiency (A, D, E, K). Bile salts are reabsorbed in the ileum and recycled. Lose that and the bile salt pool depletes, emulsification suffers, fat isn't fully absorbed — and the vitamins that travel with it aren't either. Clinically this shows as fatty, floating, foul-smelling stool.
3. Diarrhea from unabsorbed bile salts reaching the colon, where they draw in water.
4. Depending on how much is lost: general malabsorption, weight loss, and deficiencies across several nutrients at once.

If you got B12 and fat-soluble vitamins, you've understood the section. Location matters — the small intestine is not a uniform pipe, and different nutrients have different addresses.

3.5 The liver takes first look

Everything absorbed into the portal vein — sugars, amino acids, water-soluble vitamins, minerals, alcohol, most drugs — goes to the liver before it goes anywhere else. This is first-pass metabolism, and it is the reason the "detox" industry cannot survive contact with physiology.

Your liver, continuously, without being asked:

• Regulates blood glucose (stores it as glycogen, releases it, makes new glucose when needed)
• Deaminates excess amino acids and converts the resulting nitrogen into urea for the kidneys
• Manufactures bile
• Synthesizes most plasma proteins, including albumin and clotting factors
• Stores vitamins A, D, B12, iron and copper
• Metabolizes essentially every drug you take
• Processes alcohol
• Chemically modifies fat-soluble compounds — including genuine toxins — into water-soluble forms the kidneys can excrete

That last function has a name in pharmacology: phase I and phase II biotransformation. Phase I (largely cytochrome P450 enzymes) modifies a molecule; phase II attaches something to it to make it water-soluble; then the kidneys or bile clear it.

This system is running right now. It doesn't have an off switch, it doesn't accumulate a backlog that requires a juice, and it is not improved by lemon water.

🔬 Claim → Evidence → Verdict

The claim: "Your body accumulates toxins that build up over time. A detox or cleanse helps your liver and colon flush them out."

Where it comes from: Two real things, badly extrapolated. First, genuine toxicology exists — heavy metals, certain solvents, some medications really do cause harm, and genuine poisoning is treated with specific antidotes and chelation under medical supervision. Second, people often genuinely feel better on a cleanse, which is real data about how they feel — and is almost entirely explained by simultaneously stopping alcohol, sleeping more, eating more vegetables, and stopping ultra-processed food, none of which requires a product.

What the evidence actually shows: The claim is unfalsifiable as stated, because the "toxins" are never named. Named toxins are measurable and have specific medical treatments; unnamed ones cannot be measured and therefore cannot be shown to have been removed. Meanwhile the organs described as needing help — liver, kidneys, gut, lungs, skin — are doing continuous, non-optional work, and there is no evidence that commercial cleanse products improve their function in healthy people. Colon cleansing has a genuine harm profile: electrolyte disturbance, dehydration, and in rare cases bowel perforation. And several "detox" products marketed over the years have themselves been found by regulators to contain undeclared pharmaceutical ingredients.

📉 Evidence quality: No credible human evidence for benefit. Real evidence of harm for some approaches. Regulatory action on record.

Verdict: ❌ Not supported. The organs work continuously and don't accumulate a backlog. If you feel better on a cleanse, isolate what actually changed — it is almost always alcohol, sleep, vegetables, or ultra-processed food, all of which you can keep for free.

🧾 Cost check. A ten-day commercial juice cleanse of the type Theo tried in 2022 typically runs $150–$400. He described it, when I asked, as "the hungriest I've ever been," and he regained the weight within three weeks. The same money spent on vegetables he'd actually eat, over the same period, would have delivered every benefit he was chasing and none of the misery. This is a recurring pattern in this book: the intervention works, but not for the reason it's sold on, and not for the price.

3.6 The large intestine: the fermentation chamber

By the time material reaches the colon, most nutrients are gone. What's left is water, electrolytes, undigested residue, and — critically — fiber and resistant starch, which human enzymes cannot break down.

The colon does two things.

It reclaims water and electrolytes. Several litres of fluid pass through daily, most of it secreted by your own digestive system, and most of it is reabsorbed. Interference here is why gastrointestinal illness dehydrates you so fast.

It hosts an enormous microbial population that ferments what you couldn't digest. Trillions of bacteria, hundreds of species, doing chemistry your own cells cannot.

That fermentation produces short-chain fatty acids — principally acetate, propionate, and butyrate. And butyrate is remarkable: it's the preferred fuel of the cells lining your colon. Your gut lining is partly fed by bacteria, from material you couldn't digest.

This is the mechanistic core of why fiber matters, and it's a case where a mechanism and outcome data actually converge — Chapter 2's criterion for taking a claim seriously. Fiber isn't good for you because it's "roughage that scrubs you out." It's good for you because you're farming.

Transit through the colon takes far longer than everything upstream — commonly somewhere in the range of ten to sixty hours, with enormous individual variation. Total transit, mouth to toilet, is typically somewhere around one to three days.

💡 Aha moment. Notice what this does to the "detox" picture. The colon isn't a sewer holding old material — it's a bioreactor, and its output feeds your own cells. "Cleaning it out" means flushing the culture you spent years establishing. That's not neutral; it's the opposite of the intended effect. Chapter 27 takes this much further, including the parts we genuinely don't understand.

3.7 The gut is an endocrine organ

Here's the part that's least known outside the field and most consequential.

Your digestive tract doesn't just process food. It's the largest endocrine organ in your body, releasing hormones that tell your brain and your metabolism what's happening.

Two implications reshape a lot of nutrition thinking.

First, appetite is substantially hormonal, not primarily volitional. The signals telling you to eat and to stop are chemical messages generated by your gut in response to what's in it. This is Chapter 33's threshold concept arriving early, and it's the beginning of the answer to "why can't I just eat less?"

Second — and this is the strongest available evidence that any of this is real — GLP-1 receptor agonist medications work. These drugs mimic a gut hormone. They slow gastric emptying and reduce appetite, and they produce weight loss substantially beyond what previous pharmacological approaches achieved. Whatever else one thinks about them, they constitute a large, well-documented demonstration that gut hormone signalling is a major determinant of how much people eat. Not willpower. Signalling.

This is also why bariatric surgery works better than its arithmetic predicts. Procedures that rearrange the gut change hormone secretion — ghrelin down, GLP-1 and PYY up — and much of the effect appears to be hormonal rather than purely mechanical restriction. Some patients experience appetite changes within days, long before meaningful weight loss.

⚠️ When to see a professional. This section is not an endorsement or a critique of any medication. GLP-1 agonists have real benefits, real side effects, real cost implications, and genuine unknowns about long-term use and about what happens on discontinuation. That's a conversation with a physician who knows your history, not a decision to be made from a textbook or a video. Chapter 24 covers what the evidence currently supports, without promotion.

🔄 Check your understanding. Why does a large fatty meal keep you feeling full much longer than an equally caloric meal of juice and white bread? Name two distinct mechanisms.

Answer

(1) Gastric emptying (§3.3): fat in the duodenum slows the stomach's release rate substantially, so the stomach stays distended longer. Liquids and refined starch clear quickly. (2) Hormonal signalling (§3.7): fat and protein trigger CCK, and nutrients reaching the ileum trigger GLP-1 and PYY, all of which promote fullness. Juice delivers sugar rapidly with a much weaker satiety signal.

Bonus mechanism: fiber (absent in juice and white bread) slows emptying further and provides substrate for the colonic fermentation in §3.6.

This is also why "a calorie is a calorie" is simultaneously true thermodynamically and inadequate behaviorally — which is Chapter 4's entire subject.

3.8 When it goes wrong

A brief survey — each of these gets fuller treatment later, but knowing the categories now prevents a lot of misdiagnosis.

🔬 Claim → Evidence → Verdict

The claim: "Leaky gut syndrome is the root cause of autoimmune disease, fatigue, brain fog, and most chronic illness — and this supplement protocol seals the gut lining."

Where it comes from: A real phenomenon. Intestinal permeability is genuinely measurable, and it genuinely increases in certain conditions — untreated celiac disease, inflammatory bowel disease, some infections, critical illness, and with heavy alcohol use. The junctions between intestinal cells are real structures that can loosen. This is established physiology, not invention.

What the evidence actually shows: The gap is between the phenomenon and the syndrome. Increased permeability is well documented as a feature of certain diseases; what remains unresolved is whether it's a cause of those diseases or a consequence of them — and in most studied conditions the evidence points toward consequence. The leap from "permeability changes in celiac disease" to "permeability causes your fatigue, and here is a $60 monthly protocol" is not supported by human outcome data. Some individual components sold for this (glutamine, zinc, certain fibers) have plausible mechanisms and limited human evidence in specific clinical populations; none has demonstrated the sweeping benefits claimed in healthy people. Meanwhile, the diagnosis is often applied to people whose actual problem is IBS, celiac disease, or something else testable — which delays real diagnosis.

📉 Evidence quality: Strong for the phenomenon. Weak-to-absent for the causal syndrome. Absent for the commercial protocols.

Verdict: 🟡 Unclear / it depends — split down the middle, deliberately. Intestinal permeability is real and worth studying. "Leaky gut syndrome" as a marketed explanation for everything, treated with a supplement stack, is not supported. If you have persistent gut symptoms, the correct move is testing, not a protocol.

⚠️ When to see a professional. Persistent digestive symptoms deserve a diagnosis, not an elimination diet. See a physician promptly for: blood in stool, unexplained weight loss, persistent vomiting, difficulty swallowing, symptoms that wake you at night, iron-deficiency anemia without obvious cause, a family history of celiac disease or IBD, or any new persistent change in bowel habit — particularly over age 50. And critically: get tested for celiac disease before removing gluten, because the test requires gluten in your diet to be valid. That single sentence would have saved the Ortiz-Lindqvists four months and a lot of money.

3.9 How much of it actually counts?

One last thing before we leave the tube, and it's the bridge into the next chapter.

You now know that eating and absorbing are different events. So here's the obvious follow-up question, which almost nobody asks: when a label says a food contains 200 calories, how many of those calories does your body actually get?

The answer is: fewer, usually, and by an amount that varies with the food, the processing, and to some degree the person.

Where label calories come from

Food energy values on labels aren't measured for each product. They're calculated using conversion factors developed by Wilbur Atwater in the late nineteenth century:

Atwater already accounted for the fact that we don't absorb everything — his factors are metabolizable energy, adjusted downward from the raw heat of combustion. That's why protein is 4 rather than about 5.6: your body can't use the nitrogen, and excreting it as urea costs energy.

But the factors are averages across food categories, applied to specific foods. And specific foods diverge.

Where they diverge

Whole nuts deliver measurably less energy than the label states. This is one of the better- established findings in the area — controlled feeding studies conducted through USDA research have found that the metabolizable energy of several whole nuts (almonds among the most-studied) is meaningfully lower than the Atwater calculation predicts, on the order of a fifth to a quarter less for some nuts. The mechanism is structural: the fat is locked inside intact plant cell walls, and chewing doesn't rupture all of them, so a fraction passes through undigested. Grind the same nut into butter and more of the energy becomes available. Same nut. Same label. Different delivery.

Cooking increases available energy. Cooking gelatinizes starch and denatures protein, both of which make them substantially easier for your enzymes to reach. Raw starch is far less digestible than cooked; raw meat requires more chewing and yields less. The anthropologist Richard Wrangham has built an entire (contested but serious) argument that control of fire, by increasing the energy return on food, was a precondition for human brain expansion. Whatever one makes of that, the underlying digestibility difference is not controversial.

Processing and particle size matter. Finely milled flour is absorbed faster and more completely than intact grain kernels. The same wheat, in two physical forms, produces different glycemic responses and somewhat different energy yields. This is the beginning of the food matrix concept that will do enormous work in Chapter 22.

Resistant starch escapes. Some starch resists digestion in the small intestine and arrives in the colon, where it's fermented rather than absorbed as glucose — yielding less energy and feeding the microbiome instead. Cooking and then cooling starchy foods (potatoes, rice, pasta) increases resistant starch content, which is a real effect, reliably reproducible, and much smaller than the internet suggests.

And fiber returns some energy after all. Fiber isn't digested by you — but it is fermented by your colonic bacteria into short-chain fatty acids, which you absorb (§3.6). So fiber isn't zero calories; it's roughly 2 kcal/g, arriving by a different route, several hours later, having fed your gut lining on the way.

How much does this matter?

Honestly: less than the "calories are meaningless" crowd claims, and more than the label implies.

For most mixed diets, the aggregate error is probably in the range of a few percent — not nothing, but not the explanation for anyone's weight. The divergence is largest for diets heavy in whole nuts, intact whole grains, legumes, and raw plant foods, and smallest for diets heavy in refined, finely-processed foods, where Atwater factors are quite accurate. Which produces a mildly ironic result: calorie counting is most accurate for exactly the foods that are hardest to stop eating.

🔍 Why this works. Think about why intact structure reduces energy yield, because the principle generalizes. A nutrient is only available if an enzyme can physically reach it. An almond's fat is inside cells with tough walls; your enzymes can't get through the wall, only through breaks in it. Chewing makes some breaks. Grinding makes many more. Roasting and processing make more still. The same molecules, differently packaged, are differently available — and this is the mechanism behind a startling number of things: why juice hits differently from fruit, why steel-cut oats behave differently from instant, why a smoothie is not the sum of its ingredients, and why "it's the same calories" is technically arguable and practically wrong.

🍽️ On your plate. Three usable consequences. (1) Treat calorie figures as estimates with real error bars, not measurements — useful for comparison and direction, not precision. (2) Food form matters independently of food content: whole almonds and almond butter are not interchangeable, and neither are an apple and apple juice. (3) If you find calorie counting stressful or it has ever tipped into something compulsive, this section is a legitimate reason to skip it entirely. The numbers were never as precise as the app's two decimal places implied, and there are other ways to do this — Chapter 4 covers them.

3.10 The "second brain," carefully

You have probably heard your gut described as a second brain. Like most viral biology, this is one-third true, one-third exaggerated, and one-third the basis of a product line. Worth separating.

What's genuinely true. Your digestive tract has its own nervous system — the enteric nervous system — comprising a very large number of neurons embedded in the gut wall. It can coordinate motility, secretion, and local blood flow without input from the brain, which is unusual and genuinely remarkable. Nothing else in your body operates with that degree of autonomy.

The gut also talks to the brain constantly, mostly along the vagus nerve, and the traffic runs predominantly upward — the gut sends far more information to the brain than it receives. That's the gut–brain axis, and it is real physiology, not marketing. It's why nausea is generated centrally, why stress produces genuine gut symptoms, and why IBS is best understood as disordered gut–brain signalling rather than as something imagined.

Where it gets oversold is the leap from "the gut and brain communicate" to "therefore this probiotic will treat your depression."

🔬 Claim → Evidence → Verdict

The claim: "95% of your serotonin is made in your gut — so your gut controls your mood, and fixing your gut fixes your mental health."

Where it comes from: The first half is broadly accurate. The overwhelming majority of the body's serotonin is produced in the gut, by enterochromaffin cells, where it regulates motility and secretion. That's a real and frequently cited fact.

What the evidence actually shows: The inference fails on anatomy. Serotonin does not meaningfully cross the blood–brain barrier. Gut-produced serotonin and brain-produced serotonin are effectively separate pools doing different jobs; the gut's serotonin is mostly a gut motility signal, not a mood signal in transit. So "95% of your serotonin is in your gut" is true and irrelevant to the conclusion being drawn from it.

That said — and this is why the verdict isn't a flat ❌ — the gut–brain axis is real and there are plausible routes by which gut state could influence mood: vagal signalling, immune and inflammatory mediators, microbial metabolites including short-chain fatty acids, and effects on tryptophan (the precursor serotonin is built from). Human trials of probiotics for mood outcomes exist and are mostly small, short, heterogeneous in strain and dose, and inconsistent in result. Some show modest effects; many show none. This is an active field where something may well be there.

📉 Evidence quality: The serotonin statistic is accurate and misapplied. The gut–brain axis is well established mechanistically. The specific therapeutic claims rest on small, inconsistent trials.

Verdict: 🟡 Unclear / it depends — the headline fact is true, the reasoning from it is wrong, and the broader hypothesis is genuinely open. Anyone selling you certainty here, in either direction, is ahead of the evidence. Chapter 27 takes this apart properly.

⚠️ When to see a professional. If you're experiencing persistent low mood, anxiety, or changes in sleep or appetite, that is a conversation with a physician or a mental health professional. Gut health is not a substitute for mental health care, and one of the real harms of the "fix your gut, fix your mood" genre is that it routes distressed people toward supplements and away from treatment that works.

🪞 Learning Check-In

First of these; they'll appear every two or three chapters. Two minutes, honestly.

On the last three chapters:

• Chapter 2 asked you to make a Claim Filter card. Did you? If not — and most readers don't — what stopped you? Notice that the answer is almost never "I decided it wasn't valuable."
• Which of the three chapters so far felt easiest? For most readers it's this one, because anatomy gives you something to picture. Notice that easy-feeling is not the same as useful. Chapter 2 will do more for you than Chapter 3, and it felt worse.
• Did you attempt the 🧩 ileum problem before opening the answer, or read straight through? There's no wrong answer, but the retrieval literature is unusually consistent: the attempt is where the learning happens, and reading the solution feels like understanding without producing it.

One thing to try in the next chapter: answer every 🔄 prompt out loud, in a full sentence, before opening it. Out loud is the part that matters — it exposes the gaps that silent nodding hides.

Spaced Review

Three quick recalls. Try to answer before reading on.

1. (Chapter 2) What are the three questions that let you place a study on the evidence ladder in ten seconds?

Species? → Randomized? → Disease or marker? If you've internalized these, they now run automatically whenever you meet a health claim, which is the goal.

2. (Chapter 2) The "leaky gut" verdict above rests on a distinction we introduced last chapter. Which one?

Surrogate versus outcome — plus healthy-user reasoning. Increased intestinal permeability is a measured marker. The claim treats it as a disease cause. Establishing that a marker changes in a condition is not establishing that changing the marker changes the condition — the beta-carotene lesson, in a new outfit.

3. (Chapter 1) The neighbor who told Erik and Camila that Nico's problem was gluten was completely sincere and had genuinely felt better herself. Which myth shape from §1.6 is her advice using, and why does sincerity make it more persuasive rather than less?

The single cause, and to some degree restriction as virtue. Her own experience was a real observation — she is rung 3 evidence about herself. Sincerity is persuasive precisely because it removes the "who benefits?" filter: she's selling nothing, so we drop our guard. But a person can be honest and wrong, and their honesty tells you about their motives rather than about their accuracy.

Project Checkpoint: Your Digestion Log

Component three of Your Nutrition Framework. Three days, and the instruction that makes it work is counterintuitive: do not try to eat well while doing it.

The task: for three consecutive days, record what you ate and how you felt afterward. That's it.

Rules:

1. Eat exactly as you normally would. The instant you start eating "well" for the log, you've destroyed the baseline. This is genuinely hard — the act of recording changes behavior — so expect to fail at it slightly and note when you do.
2. No calorie counting. Not yet, and not necessarily ever. Chapter 4's checkpoint handles quantity; this one is about pattern and response.
3. No judgment language. Not "bad lunch." Write what you ate and what happened. If you notice yourself writing moral vocabulary — cheated, was good, indulged — circle it. That vocabulary is Chapter 33's subject, and noticing it is the first step.
4. Record how you felt, not how you think you should have felt. If the salad left you hungry two hours later, that's data, not a failure.

What you're looking for (don't analyze yet — just collect):

• Which meals kept you comfortable for three or more hours, and which didn't? Compare them against §3.3 and §3.7 — fat, protein, and fiber slow gastric emptying and trigger satiety hormones.
• Any consistent symptom after a specific food? Note it. Do not eliminate anything yet — Nico's family is the cautionary tale, and Chapter 28 explains why testing precedes elimination.
• Is there a time of day when your eating is driven by something other than hunger? No action required. Just notice.

A note if tracking is risky for you. If food logging has ever tipped into something compulsive, use the non-tracking version: at the end of each day, write two or three sentences about how your body felt, with no record of what you ate. It captures the useful signal without the mechanism that causes harm. This alternative exists at every checkpoint in this book and is in Appendix F. Using it is not a lesser version of the project.

Next checkpoint (Chapter 4): the three-day food diary — the project's core dataset, and the one Theo did.

Chapter Summary

The route, start to finish:

Six ideas worth keeping:

1. You are a donut. The gut lumen is outside your body. Eating ≠ absorbing.
2. Almost all absorption is in the small intestine, and location matters — B12 needs the ileum, and losing that stretch can't be fixed with diet.
3. Fat takes a different route — via lymph, bypassing the liver's first pass — which explains a surprising amount.
4. Your liver is detoxing right now and has no backlog. This kills the entire cleanse category.
5. The gut is your largest endocrine organ. Appetite is substantially hormonal, and GLP-1 drugs are the large-scale demonstration.
6. Label calories are estimates. Food form changes how much energy you actually get — whole nuts, intact grains, cooking, and resistant starch all shift the number, and the label doesn't know.

This chapter's verdicts:

The clinical line to remember: test before you eliminate. Removing a food before testing for the condition it might be causing can make the diagnosis impossible — and that is exactly what happened to a nine-year-old in this chapter's opening scene.

What's Next

You now know what happens to food. Chapter 4 asks what it's worth — and walks straight into the loudest fight in nutrition: "a calorie is a calorie" versus "calories don't matter."

The answer is a threshold concept, and it's going to annoy both camps: energy balance is thermodynamically non-negotiable and nearly useless as practical advice. Both. Not either. A startling number of professionals never get through that door, and once you do, most of the diet wars stop being interesting.

Theo's food diary shows up there too, with real numbers.