What if Placebos ARE the Medicine?

Season 2 • Episode 15

We’ve known about the placebo effects for over 200 years. That’s where doctors give you a pill containing no actual medicine, but you still get better. 

Recent studies have uncovered a broader range of benefits from the including alleviated pain, nausea, heart rate, hay fever, allergies, insomnia, depression, anxiety, fatigue, and even symptoms of Parkinson’s.

Weirder yet, the characteristics of the pill — color, size, and shape — influence their effectiveness.. Fake capsules work better than fake pills, and fake injections work best of all. The question is: Just how far can fake treatments go?

Episode transcript


Theme begins.

The placebo effect is when someone sick gets better by taking a fake pill, that contains no actual medicine. Or they get fake injections. Or even fake surgery. The effect is not in your head; the results are scientific and measurable. That already sounds crazy—but it gets a lot crazier.

Robson: [00:01:03] /we have these expectation effects that go kind of beyond the medical setting and to things like how easily you can perform a workout, you know, how well you can get fit, the effects of sleep loss. / even the effects of your mindset on aging and how quickly you age. These are all expectation effects that go way beyond what we once knew about the placebo effect. 

I’m David Pogue, and you’re about to hear some very strange “Unsung Science.”

First Ad

Season 2 Episode 15: What if Placebos ARE the Medicine? 

Battlefield sounds.

Every modern army employs medics to treat the wounded soldiers. And in World War II, anesthesiologist Henry Beecher was one of them.

Robson: And what he found was that when he was treating soldiers, he found that often—you know, they seemed to kind of not need pain relief when they came off the battlefield. And this got him kind of wondering whether it was just the pure relief of having been saved from the battlefield that was kind of producing its own euphoria that was then reducing the kind of pain that they were feeling. 

So they actually refused to have morphine when they were offered it. Which is incredible. And this was quite a high percentage of people. So that suggested there was some way that kind of the psychology of the soldier could influence the pain they were feeling. 

This is David Robson, author of “The Expectation Effect: How Your Mindset Can Transform Your Life.”

Robson: Now there’s this other story that says that he actually ran out of his drugs at one point. And so his nurse started injecting a salt solution instead of morphine. 

And what they found was that a lot of these soldiers really did experience that kind of pain relief. And it seemed to be quite profound. And then, you know, he studied this in later more controlled experiments and found that that was indeed the case. 

Every year, the pharmaceutical industry tests thousands of new drugs. And in clinical trials, to see if they actually work, they usually test each drug candidate against pills that look identical, but don’t actually contain any medicine. They’re just sugar pills. They’re called placebos. 

The idea is not to compare the candidate drug with doing nothing. That would be a bad study, because even if the medicine doesn’t do anything, some people will get better because of the placebo effect. The researchers might think that the medicine works, when it actually doesn’t.

No, the point of comparing the test medicine against the placebo to subtract out what’s called the placebo response. For example: If 75% of patients get better with the actual medicine, and 50% of them get better taking the placebo, you know that the actual medicine’s effect is…25%. 

Keep in mind that the people who got the fake pills might be showing improvement because they just got better over time, or saw a fluctuation in their symptoms. Either way, you now have a better idea of what effect the actual medicine candidate produces.

The best studies are double-blind, meaning that neither the researcher nor the patient knows which pills are which. That’s to make sure that the researcher’s own words and attitudes don’t somehow give away whether a patient is getting the good stuff or the sugar pill, which they worry could skew the results. 

But in the last couple of decades, scientists have started to realize: Well, wait a minute. Why are we subtracting out the placebo effect? If it’s producing positive outcomes in patients, maybe we should consider treating them with placebos.

Robson: Like if it’s providing relief, could we actually harness that effect as well—you know, potentially reducing the doses of drugs that we give to people? 

What’s crazy is that the placebo effect is not just some woo-woo, “if you dream it, it will come true” kind of thing. There are measurable biological effects. There’s been one study after another. Most of them have to do with pain.

Robson: One of my favorites was that if you give people morphine and you give it through the IV drip, so it’s kind of surreptitiously delivered, that produces much less pain relief than if you give morphine in front of the patient with the doctor telling them what they’re doing. 

In one study, researchers at Columbia and Stanford gave students free bottles of a new energy drink that contained 200 milligrams of caffeine, 2.5 times as much as what’s in a Red Bull. Or at least that’s what they told the students. 

Robson: I mean, it was just a bottle of pure water, but it was turned into a placebo. But this kind of energy drink that was meant to leave you feeling more alert and, you know, kind of energized. And that’s indeed what they found actually produced a change in blood pressure and, you know, feelings of alertness that just didn’t come obviously, when you just drink a glass of tap water. 

Pogue: Wait. Blood pressure? See, that’s another one of those examples where it crosses over from the woo-woo into the physiological. 

Robson: Yeah, exactly. / in all these ways there are objective, measurable changes. It’s not just the patient self-reporting. 

And it’s not just fake medicines, by the way. There’ve even been tests of fake surgery.

Pogue: You mentioned one study in the book where doctors would go in to put a stent in during heart surgery but not actually put one. 

[define stent]

Robson: Yeah, that’s right. The surgeon did actually kind of make a cut in their skin and, you know, like perform the actions as if they were delivering the stent. It’s just there was no stent attached to the catheter that they were inserting. The patient fully believed they were they could have been receiving the stent. It was quite a big group sample of patients, a couple of hundred. 

They found no statistically significant difference in their symptoms. 

And I should—I should emphasize actually, that this was a very particular use of the stent. So it was for treating angina. So, you know, pain in the chest. I’m not saying that doctors don’t need to use stents at all. But for the treatment of angina, it certainly seemed that there was no benefit of having the stent over just having the placebo surgery. 

Pogue: Wow. Well, the whole thing starts to make you wonder how much of real medicine and surgery is benefiting from the placebo effect. Like when we—when we buy Advil or something, I wonder if the fact that we expect we’ll get better from it is boosting its actual medical effect. 

Robson: Yeah, I think it is. And actually, you know, the drug’s marketing is really important in the size of the placebo effect. You can find all kinds of painkillers. Some of them, you know, like with bright packaging and, you know, telling you it will produce, like, ten times greater pain relief than the average painkiller. And you know, what we know is that when you have all of that positive reinforcement of the positive expectation, then that it really is more powerful. 

Pogue: And you’re saying, comparing a drug that has 200 milligrams of ibuprofen versus a generic one that also has 200 milligrams of ibuprofen? 

Robson: That’s exactly it. Yeah, you’re getting the exact same chemical. It’s just the way it’s presented. 

It gets even weirder. It turns out that bigger fake pills produce a more dramatic effect than smaller fake pills. And even the color matters.

Robson: All kinds of these things seem to make a difference. If, say, you’re trying to—um, receive like a kind of tranquilizing drug to reduce anxiety, that actually blue pills seem to be more effective in that case than if you have a red pill, because we associate blue with a kind of calmness. 

Pogue: That’s crazy. Do you think modern pharmaceutical companies know that as they decide how to design their pills? 

Robson: I expect they probably do. And, you know, that alone is enough reason for them to change the color of the pill. 

We also know that fake capsules have a stronger beneficial effect than fake pills, and fake injections have the biggest effect of all.

Now, to be clear, this stuff isn’t magic. Placebos work on some ailments much better than others. They’re amazing for pain, and do a good job on insomnia, depression, anxiety, and problems with peeing and pooping.

But they can’t shrink tumors. They don’t lower cholesterol. They don’t bring down high blood pressure. They can’t cure malaria. They can’t make a wound heal faster.

Ted: The placebo effect is primarily reducing symptoms. People say, you know, “well, medicine’s really into the underlying pathophysiology. We get rid of this chemical, fix this organ that has a disrepair, trauma,” but actually, for patients, symptoms are really important. 

Ted Kaptchuk is a professor of medicine at Harvard, where he’s the director of the Program in Placebo Studies. And what do they do there?

Kaptchuk: We do nothing but placebo research. You know, we’ve tried to find out how we can amplify placebos, make them smaller, make them higher, make the effects higher or smaller. 

Few people have been working as hard, for such a long time, to advance the idea of placebos as treatment.

Ted: So let me tell you a very simple example of a placebo effect. I could pick headaches, I could pick irritable bowel syndrome, chronic pains in the belly, but let’s start with lower back pain. 

You fall down, hurt yourself, get an injury, trauma in your back. You’re hurt. 

You’re hurt, so the doctor prescribes aspirin or Advil, maybe try some physical therapy. And in time, the injury heals, and the pain stops. For most people.

Ted: But for many people, 50 to 100 million Americans, their brain doesn’t shut off the signal. It keeps firing in the brain. That’s what most chronic pain is. Your nerves change their function and structure and they keep firing, telling you you’re in pain. 

Because we do know that when you respond to placebos, endorphins, dopamine, cannabinoids are released. Drugs hijack those same pathways. And the drugs that try to treat it, like opioids, use the same pathways as the placebo effect.

Pogue: So, you’re saying that the symptom of pain continues long after the actual injury. 

Ted: Right. Tthe brain is misfiring. It’s a false alarm. That’s very, very common. 15 to 20% of every medical kind of specialty has patients coming in saying, “You gave me this drug, I did the surgery, and I’m still in the same pain.” And doctors, what they do, they send them to psychologists. And it’s not a psychological problem, most of the time. It’s a question of how the nerves work. A lot of people haven’t caught up to that explanation. 

What Kaptchuk has discovered after 30 years of research is that there’s a lot more to the placebo effect than the pill itself.

Ted: Well, first of all, it’s not a pill, the effect of a pill, because it has no effect. It’s everything that surrounds the pill: the rituals, the symbols, the acts of kindness, the smells, the bells and whistles, of a clinical interaction. That’s really an incredible, important drama of every person that seriously wants to get healing. 

If you need any proof that the doctor’s attention really makes a difference, consider one of Kaptchuk’s most famous studies. It was a study of 262 people with irritable bowel syndrome (IBS). It’s like a perpetual stomach ache, accompanied by various pooping problems.

Ted: It’s a really nice study. It’s one of my more well-known studies, but it was published in British Medical Journal in 2008. 

He divided these people into three groups—or, as they call these groupings in the clinical-trial biz, arms. Three arms. 

Ted: We randomized 260 patients to three arms. One was, no treatment control. The second arm was, it’s a needle that looks like an acupuncture—it is an acupuncture needle. It goes in and the patient feels it. It stands there straight up. But in fact, it’s a magic sword. The needle goes up the shaft and you, you can’t tell the difference between it and real acupuncture. 

Pogue: Oh, man. 

Ted: Fake acupuncture. 

This second group had no doctor-patient relationship at all. An acupuncturist breezed in, asked if the patient was comfortable, and breezed out.

But then there was the third group.

Ted: They also got the fake needle. But then the doctor, the acupuncturist, would say, “So I’ve read your chart. I have a good idea what’s going on. But I want to hear in your own words, what is going on? How does this affect you? What symptoms are the worst, what makes it better? How does it make you feel? What things can you do that you still can do? What things you can’t do? What do you think is the cause?” 

Real schmaltzy relationship, or a little probably over the top. 

Pogue: The article about the study that I read, it said that in the third group, “practitioners were required to touch the hands or shoulders of members of their group, and spend at least 20 seconds lost in thoughtful silence.” 

Ted: You know, it was great. We went over the top. 

And we got this incredible result, as good as it can get, where the intense doctor-patient relationship, 60% of people got better. 

If they had only the paraphernalia, fake acupuncture, 40% got better. And those people who are no treatment control, they had 27% improvement. Time heals. 

And I would say that it’s clear that the patient-doctor relationship can modulate placebo effects from that study. 

Pogue: That is— so mind blowing! By a third, by a third better. 

Ted: It’s really amazing. 

Pogue: Yeah. And so, so, so this seems like a valid study. The numbers were big enough. It was well-designed. Why didn’t that instantly translate into a universe of doctors who do not rush you through the office? 

Ted: Two reasons. The evidence wasn’t good enough. But the real evidence is doctors don’t mostly care about placebo effect. I teach in a medical school, we get taught, the students get taught, that if it’s not more than a placebo, it’s worthless, right? The placebo is junk. I mean, that’s what you get taught when you’re studying development of pharmaceuticals and devices and surgery. So it’s, it’s, it’s always been tending to be marginalized, placebo. 

But the other reason that doctors don’t care about it,  is that placebos are tainted with deception and trickery,  that you had to conceal or deceive the patient and make them think it was a real drug or real intervention. So, you know, I studied with the greats, and the people that were my predecessors, my teachers and everyone believed that. 

And here’s where things get really nuts. Ted Kaptchuk is the first man who ever ran a study to answer the question, “Would a placebo still work if you knew you were getting a placebo?”

After the ad break, I’ll tell you how he tracked the answer down.

Ad Break

We’ve been talking to Ted Kaptchuk, who runs a Harvard group dedicated to turning the placebo effect into an actual medical treatment. 

But remember: Most doctors have encountered placebos only in the context of studies, where you’re giving half the patients real medicine, and the other half a fake pill that has no active ingredients. And in those double-blind studies, they tell you that you may be getting a placebo.

You could never give a placebo pill to a sick person and pretend that it’s medicine. You can’t lie to them. “First, do no harm,” right? Informed consent, and all that stuff? That would be really unethical. And so, as a result, few doctors give a placebo to patients and claim that it’s real medicine. Well, few doctors admit to it.

So Kaptchuk had what may sound like the craziest idea in all of crazy placebology. What would happen if we gave patients a placebo—and told them it was a placebo?

TED: I said, “Let’s try this. Let’s do an experiment where we tell people it’s placebo. We have to try it.” 

He approached the gastroenterologist he’d worked with on a recent experiment. 

TED: And I said, “Tony, I got to do this. I got to do this. I want to give patients placebos and tell them it’s placebos.” 

And he was great. He said, “Ted, that’s the craziest thing I’ve heard, I’m on board.” 

And I said, “we have to get a grant.” And he said, “You’re never going to get a grant!” And he said, “Ted, can you recruit the patients and pay for parking? “

I said, “Yeah, I’ll do the work for free, and we’ll sneak into the research part of the hospital and not have to pay the fine.” 

And it came out that it actually worked pretty good. 60% of people got better on the open-label plus usual care, 30% got better without extra treatment.

That blew us all away. I don’t know. It’s really hard to believe. 

In other words, compared with the people who got no treatment at all, twice as many people got better when they’d taken a placebo that they knew had no active ingredients, what Kaptchuk calls an open-label placebo or honest placebo.

But what about people who got an honest placebo versus people who got…dishonest placebos? Pills that they thought might contain actual medicine, in those double-blind trials?

Ted: Yeah, we’ve compared them directly many times. Several times. There’s no difference. 

Pogue: There’s no difference! 

Ted: I’ve published great studies with 300 people and irritable bowel, and there’s no difference between double-blind and open-label. 

Pogue: It’s so hard to believe. 

Ted: There have been now over a thousand patients randomized to this kind of thing in different conditions, like low back pain, migraine headache, knee osteoarthritis, all kinds of pain conditions, and also lots of non-pain conditions. Cancer-related fatigue, it’s been done three times now. Perimenopausal hot flashes. So there’s a lot of evidence for it. It’s hard to believe, but I—

Pogue: Yes. 

Ted: It’s because the nerves are in a context of healing. It happens automatically. It’s like the rituals, the bells and whistles, the doctor. It’s not a mind cure. It’s actually the nervous system regulates itself. It’s much smarter than the mind. The non-conscious world of our being is what’s doing it. And that’s really a breakthrough. 

Pogue: But it occurs to me that there would be no side effects, either. 

Ted: No. If a person is honestly told it’s a placebo, there’s no side effects. 

Pogue: When I was a kid, and I found out placebo effects about the same time I found out about this effect, that the placebo effect still works if you know it’s a placebo. 

Ted: Yeah. 

Pogue: You’re the guy who figured that out?!

Ted: Yeah, that’s my first experiment, and I’ve done 15 of them since then. 

Pogue: That was you?! That, that has been a cornerstone of my cocktail party conversations— 

Ted: I really want to take that taint out of placebo. The idea that you need deception. In some ways, healing rituals tricks our minds to turn down the amplified pain sensation. It turns down the false alarm. 

Pogue: Absolutely incredible. So do we have any idea what the physiological explanation is for that? 

Ted: Oh, thank you. Thank you. Thank you. It’s a really important question to ask. 

We don’t 100% know everything what’s going on there. But, the simplest explanation is that the nervous system amplifies symptoms, not only pain. And sometimes it gets stuck there, and up in the brain, the pathways that make the pain light up, the false alarms, are the same pathways that placebo sometimes, in some people, turn down. 

We know from—for a long time, that for double-blind placebos and deceptive placebos, neurotransmitters were involved. Like, like endorphins—if you respond to placebo, the release of maybe endorphins, cannabinoids, dopamine. And what we now know, recent experiments tell us, that even when it’s open-label, you get the endorphins involved, right? 

So we know that, in many cases, after you take a fake pill, your brain releases real chemicals, which produce genuine improvements in your symptoms. We know that the placebo effect gets magnified if you get a lot of attention from the doctor. Schmaltz, as Kaptchuk calls it. 

Pogue: Okay, so in three groups in general: a group that gets no care might get less pain over time, just because it fades. Placebo effects with no schmaltz does better than that. 

Ted: Right. 

Pogue: But a placebo pill plus schmaltz—

Ted: —will get a bigger placebo effect. Yeah, yeah. 

Now, remember our author friend, David Robson?

Robson: I’m David Robson. I’m 35 years old. I’m the author of “The Expectation Effect.”

There’re actually only two chapters in his book about the placebo effect. To him, the placebo effect is only one form of the expectation effect. Placebos are a subset of the effect that expectations can have on your life. 

Robson: One of the best examples concerns exercise. They gave these students a genetic test for a variant that’s known to affect your kind of capacity for endurance exercise. So if you have one variant of the CREB-1 gene, it seems that you are a bit, you know, better able to do endurance exercise. And that’s reflected in physiological measures like the gas exchange within the lungs. 

So they gave these students this genetic test, but then they gave them sham feedback. So the students didn’t initially find out, you know, what variant they had. 

And what the researchers found was that those expectations alone, independent of the genes they were carrying, influenced their performance in this endurance exercise. And in some cases, the influence of the expectations was actually greater than the influence of that gene. So for the gas exchange within the lungs, the expectations were a bit more powerful. 

Pogue: So if I told you, “oh, lucky you, you’ve got the gene, you’ll—you’ll run better, longer”—you believe it and you do? 

Robson: Yeah, exactly. That’s what they found. And, you know, it also affected feelings of how hard they were working out. If you thought you had the good gene, you could be really like going for it on the treadmill, but it didn’t feel like so exhausting.  

So I think it speaks more broadly to, you know, the narratives we carry around us. Like, if you’ve always just assumed that you just aren’t cut out for exercise, maybe because of—you know, what you see in your family members, maybe because of, you know, memories from high school of not really enjoying gym class—well, that is actually going to have a similar kind of expectation effect on you. 

And then there’s the aging study, performed in 2002 by Becca Levy at Yale.

Robson: So that’s the one that really blew my mind. She found that people—people’s kind of self-reported expectations of aging, you know, at midlife seem to predict their longevity by seven and a half years. 

So if they thought that things would get better with age, they lived longer. If they expected that things would automatically get worse with age, they lived seven and a half years less than the other people. So a huge effect. 

That includes getting the various diseases you can get when you’re older, too. Here’s Becca Levy herself, in an American Medical Association video:

Levy:  We were able to look at people who had the riskyou gene for developing dementia. And we were able to look at people who were free of dementia at baseline, and then we followed them over six years to see whether they developed dementia. And we found that even in this high-risk genetic group, if they take in more positive age beliefs, they actually had, we found, they had a 40% reduced risk of developing dementia. Their risk of developing dementia was as low as people who were not who are not born with that risky gene. 

Robson: There’s a really big body of evidence that’s built up over the last 20 years that shows this to be the case. So the big question is, well, how could that happen? 

Well, one explanation is that if you think things get better with age, you take better care of yourself. But Robson says there’s more to it than that; the expectation effect messes with your release of damaging stress hormones.

Robson: So what you see is that people who have the negative expectations of aging, they start to feel a lot more vulnerable and they’re going to see the challenges around them. Even something like going to post a letter, you know, going to the store to get your groceries, you’re going to start worrying about, you know, losing your way or having a fall. And that sense of vulnerability increases the kind of stress that they’re feeling. 

So you see for these people, a steady rise in the stress hormone cortisol. And again, that then causes a steady rise in inflammation, which we know causes kind of bodily wear and tear. 

And over time, you know that the consequences add up, and it just puts you at a higher risk of all of these different illnesses associated with aging, and eventually your mortality. 

Pogue: Well, what should we do with that information? Should medical science at every checkup tell you getting older is nothing to worry about? 

Robson: Right.  I think we do need to take action. And, you know, Becca Levy in her first paper was like if we found that there was some kind of virus that was reducing people’s longevity by seven and a half years, we would be taking action. But actually, what she argued is that, you know, the ageism that permeates our culture is, you know, is a pathogen that is having that effect. 

We as a society—we should be fighting ageism a lot more fervently than we do at the moment. We need to stop reinforcing this message that as you get older, it’s automatically a time of vulnerability and decline. 

So I think that’s a no-brainer, really, is that we have to kind of, you know, be a bit more careful about kind of what messages we’re sending to other people and ultimately what messages we’re internalizing ourselves. 

You know…I watched a couple of Becca Levy’s videos. At one point she suggests keeping an ageism journal.

BECCA: What that involves is, for one week, writing down all the messages about aging that you encounter, whether it be in social media, whether it be in a magazine, seen advertisements, whether it be, you know, talking to a relative or overhearing a conversation in the coffee shop. Write it down, and then write it down whether it’s positive or negative. And if it’s negative, take a moment and think, “could there have been a different portrayal of that older person?”

You know? It’s true. Think about it:  It’s no longer cool to make fun of people’s looks, or race, or disabilities. But for some reason, making fun of old people is still fair game. 

Dyer: The conversations I had with my dad at the end of his life were the same ones we had when I was about six; the roles just flip around. We would go out to eat, walk into a restaurant, I’d look at my dad and go, “Listen, sit right here. Don’t touch anything, don’t talk to anyone. I’ll be back in a minute, all right? Where are your shoes?”

That’s comedian David Dyer on YouTube. Sure makes me look forward to getting old.


So what have we learned? The placebo effect is real, it’s measurable, it produces physical changes in your body, and it’s freaking weird. Placebos even work when you know you’re getting them. Even though you know they can’t work on you, they work on you.

If you ask Harvard’s Ted Kaptchuk, modern medicine is just ignoring a vast realm of potential treatments that could be helping people—right now. He believes that placebos shouldn’t just be a nuisance variable in medical trials; placebos should be considered treatments in their own right. They work!

But making them mainstream will be an uphill battle.

Ted: But it hasn’t caught on. And I ask many of my colleagues who’ve been doctors in my, in my studies, I say, “Why don’t you use it when you’ve got these great results?” They say, “it’s not standard of care, Ted.”  You didn’t get trained in medical school to give placebos, you’re trained not to give placebos. I give them a break. And it’s going to take a shift. 

But, you know, I never stop fighting. So let’s see where it goes.

The Man Who Invented QR Codes

Season 2 • Episode 14

In 1994, Masahiro Hara, working at a Toyota car-parts subsidiary, got tired of having to scan six or seven barcodes on every box of parts that zoomed past on the assembly line. Why, he wondered, were we still using the standard barcode—a bunch of closely spaced parallel lines—that we’d been using since the 70s? Why couldn’t someone invent a barcode that used two dimensions instead of one… could work from any angle or distance…could work even if it got smudged or torn? 

And so, studying a game of Go, he dreamed up what we now know as the QR Code, the one you scan with your phone. It’s the square barcode that shows up on restaurant menus, billboards, magazine ads—even tattoos and gravestones. But even that, says Hara-san, is only the beginning.

Episode transcript


Even if you don’t know what a QR code is, you actually do. You’ve seen it hundreds of times. It’s a printed square made up of black-and-white square pixels in weird patterns. They show up on ads, business cards, tickets, restaurant menus… You point your phone’s camera at it, and boom—it opens up a website, or a menu, or a show ticket. 

Pogue Do you remember the day that you came up with this idea? How old were you? Where were you? 

Translator Yes, I remember very clearly. It was early 1993, when I was 34 and when I was playing Go, which is a board game, during a lunch break. The concept of putting black and white dots on a grid occurred to me. 

It’s rare to find a cultural element as global and ubiquitous as a QR code…that was basically invented by a single guy. And today, you’ll get to meet him.

I’m David Pogue, and this is “Unsung Science.” 

First Ad

Season 2, episode 14…The Man Who Invented QR Codes!

Yes, that’s right. I’m going to devote an entire episode to the invention of a barcode. 

I do realize that if I really want to soar to the top of the podcast charts, this is not the right topic. I should do true crime, or partisan politics, or answer your sex questions. But you know what? I believe in myself, and my interests. I have integrity, and so do you. So…barcodes it is. 

You’ve seen thousands of barcodes in your life. On every single thing sold in every single store. Every bottle, box, bag, can, carton, container, crate, jar, jug, packet, pouch, pack, pallet, sack, and tube… has a barcode.

It’s that little patch of thick and thin lines, stripes, like unevenly spaced fence posts or jail bars. The cashier scans that barcode, or you do, and presto—

Beep! of a checkout scan

…the cash register knows what item you’ve bought and what the price is. And the store now knows that you’ve depleted its inventory of that item by one. 

In the pre-barcode days, the cashier had to look for a price tag on every single thing, and manually type the price into the register…

Cash-register tabulation sounds

…which was tedious and error-prone and gave you carpal tunnel syndrome.

Cash-register tabulation sounds, followed by “ow.”

What I’m describing, of course, is a UPC barcode. Universal Product Code. It came along in the 1970s, when the grocery-store industry decided that they’d had quite enough of the old—

Cash-register tabulation sounds, followed by “ow.”

It was time to modernize. They opened up a competition to see which tech company could design the best scannable barcode, and in 1972, IBM was the winner. It designed this barcode based roughly on Morse code, with its alternation of short and long tones…

Morse-code sound

…only this time, it was thick and thin lines read by a laser beam.

And in the summer of 1974, the first store rang up the first sale of the first item ever bought by having its UPC barcode scanned. 

Beep! of a checkout scan

It was a pack of Juicy Fruit gum, for 67 cents—a purchase so culturally significant that there’s a replica of that pack of gum in the Smithsonian. 

Now, at checkout, the UPC code was much faster and more accurate than human eye-hand coordination. But as human-computer interfaces go, it was really pretty crude. 

Problem #1:


These barcodes are one-dimensional. If the code isn’t perfectly perpendicular to the laser beam reading it, you get an error. Of course, IBM’s design features two laser beams, forming an X, inside the scanner, so you can be a little sloppier with your barcode positioning. But still, it sometimes takes a few tries to get it to read.

Problem #2:


…was tolerance for errors: the UPC didn’t have much. So if par t of the barcode is torn away, or smeared, or partly covered by a Sharpie mark or an oil stain, the machine can’t read it.

And problem #3:


The standard UPC barcode doesn’t store much data. Its entire readout, in fact, is 12 digits. And the biggest number you can describe with 12 digits is twelve 9’s in a row—just shy of 1 trillion. It’s 999 billion, 999 million, and so on.

Everything sold in every single store has its own unique UPC number. A 20-ounce bottle of Heinz ketchup is always 013000013673. For half-inch Scotch Magic Tape, it’s 021200000041. So in theory, the UPC system should be able to identify a trillion different products, right?


Turns out the first six digits identify the company. The barcode for every product from Procter & Gamble, for example, begins with 003800, whether it’s Febreze, Pampers, or Tide.

Then the next 5 digits identify the product within that company’s portfolio. Procter & Gamble, for example, makes Joy Lemon Scent Dish Soap, 21737.

Now, if you’re following along at home, you may realize that we’ve accounted for only 11 digits. Six for the company, five for its product. What about the twelfth number in the barcode?

That’s a checksum—a digit that confirms that this is a real UPC code, and that it’s been scanned correctly. This is super wonky, but I’m going to get into this, because it’s kind of fascinating: To see if a UPC code is real, you add up the numbers in the odd-numbered positions and multiply by three. Then you add in the digits in the even-numbered positions. Divide your answer by 10, subtract that answer from 10, and voila: You’ve just calculated the final digit of the UPC code. The checksum. 

Credit cards use a checksum system, too. That’s why you can’t make up a credit-card number.

Just for fun, I looked up the barcodes 111111111111 and 222222222222. Those turn out to be invalid codes, because the checksum comes out wrong. 555555555555 works, though. That’s an actual product. It’s an L-shaped pillow for breastfeeding babies called the Jolly Jumper Boomerang Nursing Cushion. 

See what you can learn from podcasts?

Anyway, bottom line is that UPC codes will run out after far fewer than a trillion products. The limit is more like a few billion. And we’re getting to that point. Have you seen how many flavors of Oreos there are?

Over the years, various UPC code systems with expanded capacity have come along. But all those barcodes are still one-dimensional, error-prone, and limited. 

Which is why this episode is not about UPC barcodes. It’s about their spiritual successor: The QR code.

Now, you may or may not know the term QR code, or what it stands for: “Quick Response.” But you’ve certainly seen a QR code. It’s that square computery pattern, made up of little square black-and-white pixels, that shows up on things like billboards, magazine ads, business cards, tickets, and restaurant menus. You show it to your phone’s camera, and it does something. It opens a web page with more information, or shows you the restaurant menu. You might have used a QR code to check into a hotel, or log into a web site. A QR scan can pinpoint a location on a map, display a message or a picture, download a PDF file, auto-connect to a WiFi network, or pay for something.

QR codes appear in books. They show up on baseball tickets in Japan. You even see QR codes during TV shows or TV ads, to scan from the couch. 

In some countries, during the pandemic, you had to scan a QR code at stores and restaurants to show that you were there, for contact-tracing purposes. In other countries, vendors have set up entire virtual stores in train stations—basically a wall of photos of items you can order on the spot, by scanning their QR codes.

Nigeria, Russia, and the Netherlands have released bills or even coins with QR codes on them, which you can scan to read up on some historical national info.

In China and other countries, QR codes serve as sort of interactive price tags: You scan the QR code for an item, and your phone says, “Pay 12 bucks?” or whatever—and with one tap, you’ve paid. This system is so fast and easy that almost nobody uses cash anymore in China. Stores, cabs, subways, movie theaters, street vendors, street performers, even people experiencing homelessness have QR codes for easy paying. 

So yeah. QR is everywhere.

So what is this thing? Where did it come from? And who invented it? And is he a multimillionaire?

Well, we found him. He’s alive and well, in his mid-sixties, living in Japan, surrounded by shelves full of awards.

Pogue When you go to parties, when you go to meetings, do people know who you are? 

Translator Yes, I believe so. Especially with these days after and the honor of winning multiple awards, people started to recognize my achievement of developing QR code. So sometimes I was asked to take a photo. So gradually, yeah, people recognize. 

This is the voice of Masahiro Hara, the man who invented the QR code. 

Actually, it’s not. This is the voice of Masahiro Hara:

Hara: Same response in Japanese

But in an “Unsung Science” first, I conducted this interview with him over Zoom, with his colleague Yoshihiro Okamoto serving as translator. 

Hara-San was born in Tokyo in 1957. Yeah—I’m going to refer to Masahiro Hara as Hara-San, because in Japan, that’s how you refer to someone with respect. “San” kind of means “the honorable” or “good sir.”

So. Hara-san graduated college in 1977, and got his first job at Denso, a Japanese car-parts manufacturer, a subsidiary of Toyota. He was still at Denso in the early 90s, when the Japanese economy wasn’t in great shape. Every company was trying to boost productivity. 

And you wanna know what was extremely unproductive? Trying to find the right box of parts on the Denso assembly line. Every box scrolling by had multiple UPC-style barcode stickers on it. You’d have to scan every one of them with your handheld scanner gun to figure out what was inside. It was super inefficient, and also a pain in the butt.

Translator There were cases where we put ten barcodes side by side, and read it one by one. So it’s very inefficient at that time. 

Hara-san was 34 years old, and he’d had enough of scanning eight or ten stickers on every box of car parts. 

Pogue Did somebody ask you to create a new barcode? Was that an idea from your boss, or did you just independently say, “I think we need a better barcode?”

Translator The idea occurred to me on my own. 

Pogue Do you remember the day that you came up with this idea? How old were you? Where were you? 

Translator Yes, I remember very clearly. It was early 1993, when I was 34 and when I was playing Go, which is a board game, during a lunch break. The concept of putting black and white dots on a grid occurred to me.

It may have been the most important lunch break of his life.

Go is an ancient Chinese board game where you and your opponent each have these round, white or black stones. They look kind of like Mentos, or oversized M&Ms. The object is to place them strategically on a 19 by 19 grid in such a way that they fence in your opponent’s stones.

If you saw a Go board after a game, pictured side-by-side with a QR code, you’ll definitely see the family resemblance.

Anyway, the key to this Eureka moment was that UPC barcodes are one-dimensional. If the code were square instead, two-dimensional, you’d be able to store so much more data.

So Hara-san told his bosses that he wanted to develop a new, improved scannable code for Denso’s car parts. As translator Okamoto-san puts it,

Hara-san: When he decided to develop this, he told his boss that that he’ll do it in two years.

Pogue Wow. How long did it take? 

Translator Exactly two years. 

The first challenge was figuring out how the scanner camera would know when it was looking at a QR code—to differentiate the code from whatever text surrounded it on the page or the box. How could he teach the software to pluck out the QR code from its surroundings?

And then, one morning on the train to work, buildings were flashing by. Façade after façade, each full of windows in identical rows and columns. 

But suddenly, one building jumped out at him: In a bit of whimsy, the architect had designed the windows at the top and bottom of the building to be different shapes and sizes. 

Maybe that was the key. Maybe he could put special locator symbols, finder symbols, at the corners of his code that would tell the scanner, “Yo! Start interpreting this as data!” 

But what would those finder symbols be?

Translator So in order to distinguish QR codes from letters and figures, I investigated various printed materials to find shapes that are rarely used in the work at that time. 

It had to be a really unusual symbol, something that would never ordinarily appear in printed material. Otherwise, the scanner might try to read something that was unreadable. 

Hara-san studied fliers, boxes, newspapers, magazines, and books, in multiple languages, trying to find something that wasn’t there—a symbol that nobody was using. 

What he came up with was simple and elegant: A solid black square, with a black frame around it, like a little cubist eyeball. A square inside a square. Turns out that symbol almost never appears in any other context. It’s not part of any alphabet.

If you look at a QR code, you’ll see three of these finder symbols, at three of the four corners of the square barcode.

Pogue Why three and not four? 

Translator Because QR code is a square, so if, you know, the—two lines you can define what is the other one, where the other one is, where the other one is. 

In other words, if the software knows where three of the corners are, it can figure out where the fourth one is.

 Pogue Did you try any other designs that didn’t work well? Other patterns, other shapes? 

Translator Yes, there are some other ideas, especially for making a finder symbol. And for example, there was an idea to make a triangle or circle in the corner. 

Pogue But square worked better? 

Translator Yes, square worked better. 

Once the computer knows that it has encountered a QR code on the page, and knows where its boundaries are, it knows the code’s orientation, and it can begin to read the actual data—that ocean of tiny square pixels. 

Translator Okay, so a scanner, which is a camera, first takes image by its camera. Then it recognizes the finder pattern, as we discussed, in the QR code. And then from the three finder patterns, the scanner identifies the outline of the QR code and reads the black and white pattern of each cell. Then finally, it shows the characters contained in the code, and it is done in 16 milliseconds. 

The software starts reading from the lower right, where the very first cluster of dots tells the software whether this message is going to be a number, some text, or some Japanese characters. The next cluster of dots specifies how long this message is going to be—how many numbers or characters. 

The analysis zigzags through the rest of those pixels like a tractor mowing a field: It scans upward until it hits the first cubist eyeball, then turns around and scans down the next column to the left. Hits the bottom, doubles back and scans up the third column, and so on. 

Eventually, it encounters a special cluster called the “end indicator,” meaning, “This is the end of the message.” But the scanning isn’t over yet. At this point, the path continues with error-correction data—kind of like that checksum digit in the UPC code, but much, much more detailed. It’s almost like a backup of the QR’s encoded data.

All told, a single QR code can store over 7,000 numbers, or about 3,000 typed characters. The grid of dots can be as small as 21 rows and columns, or as big as 177 by 177.

The real genius of the design, though, is not just the amount of data it stores; it’s how fast and flexible it is for reading.

Translator QR code can store 200 times more information than barcode, and it can be read from anywhere in 360 degrees quickly. And also it can be read even if part of the code is damaged or covered with dirt, which barcode cannot. 

In fact, even if you smudge or cover or tear away as much as 30% of a QR code, the information still comes through to your phone. That’s how much error-correction and redundancy is built into the design. 

Sometimes, companies dress up a QR code with their logo, or some little graphic in the middle. I saw that on a cereal box recently. They can do that because of this error-correction. The scanner says, “oh dear, there’s a chunk of this code covered up by some graphic—but I don’t care! I can still get the data!”

You can be amazingly sloppy with the angle or distance of your phone. That made it a huge hit in the Denso car-parts factory. You could reliably scan boxes that had weird shapes, sitting at odd angles as they zoomed by on the conveyor belt. 

After two years of effort, Masahiro Hara had achieved his goal.

Pogue Were your bosses very happy? Did you get a raise? Did they have a party for you? 

Translator Actually, my bosses were not so excited when I first showed them the QR code. It Is because they did not know how much it would be used, or generate new business with this new code. 

Well, great. Two years of genius effort, and Hara-san is rewarded with a big yawn from management.

Obviously, there’s more to the story. In particular, there were four seismic cultural events that changed the course of history—and brought QR codes to global domination. I’ll let you know about ‘em…after the break!

Second ad break

Before I get back to QR codes, I gotta tell you something really funny.

After we record an interview for this show, we feed it into an AI transcription website called Trint to convert it into a written document, so I can write my script.

The accuracy is not flawless. So I hire a wonderful person named Oli via Noble to listen through the interview and fix the Trint transcript.

Now, Trint is capable of transcribing 33 languages—but you can choose only one language per recording. My conversation with Hara-san included both Japanese and English, so I chose English and clicked Transcribe.

WELL! Trint did its best trying to transcribe the AUDIO of Hara-san’s answers in Japanese—into English syllables that SOUND like what he said, even if they make no sense as sentences.

So, for example, when Hara-San said:

Hara: [00:27:39 (excerpt)

…Trint treated him as an English speaker saying this:

Olivia: Oh. And of course, they got like Scott Bakula crying, also looking to get them.

That’s Olivia herself, reading these comic gems. Here’s another one:

Olivia:  So we just got a kick out of my Yoda knuckles. I could not shake your head on.} 

And who doesn’t get a kick out of Yoda knuckles??

OK, OK, just one more. 

Olivia:  I’m sure, according to psychologist and Uncle Sam, somebody just took our idea based on all the dictum, one of the most radical terminology in the Middle East.

So great. Well done, Olivia. 

OK. That was just a little comic interlude for ya. Now, back to our story in progress.

Hara-San had indeed licked the problems of traditional barcodes. But his bosses failed to appreciate the QR code’s genius and its majesty. What they wanted to know was, how’s it going to make us money?

Translator They did not know how much it would be used, or generate new business with this new code. And so they told me that you should go to the market first, then see how, you know, potential customers respond. 

So Hara-san took his invention to auto-industry trade shows and gave demos to potential clients. 

Translator Fortunately, we got a very good response from the potential customers. So that makes, you know, my bosses’ response gradually different. And there was a party for me six years after the code was invented. 

Pogue That’s a little late. 

Once the Denso executives saw that the QR code had money-making potential, they made a crucial decision: Give away the technology. Don’t defend the patent. 

Pogue Why did Denso make the QR code format available to the whole public? How would the company profit from the QR code becoming popular? 

Translator Denso’s company strength is manufacturing. So our strategy was to develop business with readers, code readers, for QR codes. 

Pogue Oh, I see. So you give away the code and then you charge for the reading machines. It’s like, like razors and razor blades, right? 

Hara-san Yes, I think so, too. It’s a similar model. 

Pretty soon, all the car companies were using QR codes. Then advertisers. Then the Japanese lottery. The QR code was a surprise hit—in Japan.

Pogue And by the way, we call it QR code for ‘quick response.’ Is it a different name in Japan? What did you call it? 

Translator In Japan, it’s also the same—quick response code. So we call it QR code too. 

But becoming a cultural triumph in Japan is one thing. It took decades for the QR code to achieve full world saturation. And it owes that ubiquity to four seismic cultural events.

First, the mad-cow disease scare. Don’t know if you remember that, but it was this horrifying neurological disease in cows that made them uncoordinated, nervous, or even violent, and then they died.

Anchor: In 1996, a lethal new disease appeared in Britain. 

Doctor: Patients present with difficulties in balance and walking. And the deterioration from first symptom to death takes only a matter of months.

In the nineties, a couple of hundred people died after eating contaminated beef, so there was a worldwide panic. Suddenly, it was really, really important to track every bite of beef, from the farm to the wrapped package in the grocery store. The QR code was an ideal tracking system.

The second big push came along in 2007, with the invention of the iPhone. Pretty soon, people could download a special app onto their smartphones, and read QR codes just by pointing the phone. You didn’t need to buy a special reader gun anymore. Sorry about that, Denso!

The third huge event: In 2017, both Apple and Google built QR scanning right into the Camera apps of their smartphones. No more downloading a special app!

Pogue Somewhere along the line 2017, the phone could read QR code just by itself, just in the camera app. Did they call you up and ask you about that? Did you know that was happening? 

Translator No, actually, they did not consult with us. But when this happened, I felt very happy, because I thought the QR code has been widely recognized all over the world. 

Today, you just open your Camera app and point it at a QR code. You don’t even take the picture. Instantly, a button appears, showing the website that will open when you tap it.

And the fourth push was a little thing called the global pandemic. During COVID, nobody wanted to handle a restaurant menu that might be infectious, or to pick up a brochure or whatever. Everything went touchless. The solution was—QR codes. Point your phone at a laminated card on the restaurant table, and the latest menu or wine list appears on its screen. Saves the restaurant money, makes it easy to update the menu, and keeps your grubby mitts off of physical menus.

Suddenly, QR codes went mainstream. 

Today, Masahiro Hara is still working at Denso after 46 years. And he’s still developing the QR code’s successors. Like, what if you could wear a QR code on a bracelet or necklace with all your medical records? 

Translator What I am working on now is to put information about someone’s X-ray data or heartbeat data. So when someone goes to the hospital, if he or she can show that kind of information, he can or she can have a smooth and quick diagnosis in the hospital. 

He’s experimented with color QR codes, too, going beyond black-and-white to pack even more data into a tiny space—maybe even videos.

And he’s already introduced the SQRC—the secure QR code, where part of the code is encrypted, and can only be read by a special scanner.  

Translator: As a result, it is used to poll amusement park tickets and also for traceability. 

Of course, once you’ve got a secure portion of the code, that can’t be faked or duplicated, all kinds of possibilities open up. For example, you know those employee security badges that get you into secure areas of a building? Hara-San has come up with a way to make them un-impersonatable, using those same SQRC’s.

It works like this: Your employee photo is embedded into the encrypted code on your badge. Now you show up for work, at the nuclear facility or gold-bar storage company or whatever.

Translator: To verify the person, he or she goes in front of the camera and holds SQRC held up by the dedicated reader. 

At this point, a camera compares what you actually look like, standing there, with the stored image of your face on your badge. And if they match, you get to go in.

Translator: You can even use this system in an environment not connected to the Internet. So as a result, this system is used for access control, or it can also prevent impersonation. 

Pogue Your QR code obviously became very successful. Was that because of your skill and your good ideas, or was there some part of luck and timing? 

Translator It was my idea to develop a code that can be read quickly and reliably. But I was very lucky to be able to develop this before a cell phone or a smartphone with cameras become popular. 

Pogue I’ve heard of QR codes appearing on gravestones, which seemed surprising, but also a very good idea. Have you heard of any crazy examples of people using QR codes?

Translator Yes, I have seen people with QR code tattoos on their bodies in Japan in 2004. Then I heard their contact information in in there, so they exchanged their contact information by using a tattoo. 

That seems like a pretty cool idea—as long as you don’t think you’ll ever move or change your phone number for the rest of your life. Oh wait a minute, no—if your address changes, you’d just edit the website that opens up when you scan the QR code, like a restaurant changing its prices. Duh.

Pogue Did you have an idea that this could be very big? 

Translator The answer is yes and no. So when I went to the trade show and showed, showed especially industry-use customers this secure code, I had a very good response. So I thought it can spread in an industry use, but I could never imagine that it would be spread, you know, you know, usual consumers. So that was totally surprising to me.

Heroic, sunset music

I’m delighted to have met Masahiro Hara—truly one of the unsung engineers of our time—and to tell you his story. And when you look back over the arc of his career, I think one profound adage really sums it up. In the words of the Trint transcription:

Hara-san:  [00:21:13] But the real concrete sketches on display? Your eyes.

As I said goodbye to Hara-san and his translator Okomoto-san, Hara-san had a little surprise for me. 

Pogue [00:48:36] /you know, young people today, they might have posters on the wall of their sports heroes or their music heroes. If I had a poster on my wall, it would be of you. 

Hara-san [00:48:21] Thank you. Thank you for interview me.

That’s Hara-san himself, speaking his own English!

Pogue [00:48:25] Oh, you didn’t tell us you speak English! 

Translator [00:48:33] He can. 

Hara-san: I am happy that you are interesting in QR code.