I’d like to announce the newest addition to the Mackenzie book collection! What’s Happening in the Mathematical Sciences (WHIMS) is an ongoing series published by the American Mathematical Society, about… well, just what the title says: current and recent events in mathematics. This is the eleventh volume in the series, six of which I have written or co-written. The nine topics I chose for this volume were:

• The mathematics of gerrymandering
• Mathematical models of dieting, and why it’s so hard to keep the weight off
• The detection of gravitational waves, and how math made it possible
• Topological data analysis, a step beyond simple clustering
• The Cap Set Conjecture, or how the game of Set met its match
• Mathematical models of bike-sharing and ride-sharing networks
• Expander graphs and their applications
• Quantum computers, and specifically the search for “golden gates”
• The simplest equation in math (x + y = 1) and how it has led to new progress on the asymptotic Fermat’s Last Theorem

As you can see, I tried to cover a broad spectrum of applied and pure mathematics. The chapters also have a range of difficulty, from topics a high-school student can easily understand to some that are at the level of a college math major. However, even with the more difficult topics I try always to show the “big picture” of why it matters, and I try to stick to Einstein’s dictum, to make it “as simple as possible, but not simpler.” The book has many color illustrations, which I think are part of the series’ appeal.

WHIMS volume 11 does not appear to be available yet on Amazon, but it is available through the American Mathematical Society’s website, $25 for non-members and $20 for members. 

This book is a bit of a personal landmark for me, as it is my tenth published book. I never expected to get to double digits! The significance of this landmark dawned on me recently, when I decided to compile a list of all the book published by graduates of the Science Communication Program at UC Santa Cruz. It was quite an eye-opening list. At the end I tallied up a “leaderboard” of the most prolific alumni book writers, and I was quite surprised by who came in #4.

1. Emily Sohn — 57 (!!) books published
2. Mike Wooldridge — 37
3. George Wuerthner — 29
4. Dana Mackenzie — 10

While I have no chance of ever catching the overachievers who are ahead of me, it made me feel pretty good to realize that out of 350+ graduates in the program’s 38-year-history, only three have published more books than me. By the way, I strongly recommend Emily Sohn’s books! She writes pictorial books or cartoon books on science for youngsters. I have to think that some of those 57 books may have changed lives, by getting kids excited about science at an early age.

Finally, here’s one more look at my latest book. All of my books have to pass the sniff test!

Although I recently had a small reminder of the value of humility, today I’m going to soldier on in my immodest way and celebrate the first anniversary of one of the most exciting days of my life, the publication day of The Book of Why.

Being a writer, especially of books, is a strange business. You have to spend a very long time, months to years, working on a project with almost no encouragement or validation. It’s just you and your editor, and your co-author if you have one, as I did. You have to believe very strongly in what you are doing, because for the longest time that belief is the only thing that sustains you.

Then comes publication day, and suddenly the validation is very visible and very public. It’s like emerging from a cave into the dazzling sunlight. It may not have always been this way, but nowadays we have Amazon, which gives a writer immediate, hour-by-hour feedback on how the book is selling. Not actual sales figures but a ranking, which to a ratings junkie like me is like an opiate. (I used to religiously listen to American Top 40 every week as a teenager.)

A year ago today I got online at Amazon and checked my ranking. #5027. I was actually kind of disappointed, because it had been in the 4000’s the previous day due to pre-publication sales, and so it didn’t look as if publication had actually made much difference.

But then the fun began! By noon we were up to #3076, and my editor e-mailed congratulations. By 1 pm, #2179. By 5 pm, when I came home from the Aptos Library chess club, my wife had taped to the door a note saying “#1296!”. And then at 8 pm, we cracked the top 1000, at #952, something I had hoped and dreamed of but never really thought would actually happen.

Although that day was thrilling, for the most part that isn’t what the book business is about. What really matters is the next 364 days… and the 365 days after that… and so on. Thank goodness, the book business is the ultimate long-tail business. The value of a book is not measured by how many copies it sells on the first day. Thomas Kuhn’s The Structure of Scientific Revolutions sold less than 1000 copies in its first year.

Over the last year, the reviews have been pretty good, the sales have been pretty steady, and Judea and I keep seeing encouraging signs that our message has been heard. For example, just a week ago he sent me an article about the head of artificial intelligence at Toyota, who basically quoted straight from our Chapter 1. We had written about something called the ladder of causation. The Toyota guy renamed it as “the ladder of autonomy,” which I think is an even better name within the context of AI. This is great! When people are taking your ideas and re-thinking them, it really shows that you are getting somewhere.

Of course this success is 99 percent due to Judea. I have no illusions about that. I am just enjoying being along for the ride.

Anyway, this morning, just for old time’s sake, I got onto Amazon and checked our book’s ranking. #6546. It won’t be zooming up into the 3000’s or 2000’s or 1000’s today. But after a whole year it’s still #1 in the “Discrete Mathematics” category and therefore gets the little “#1 Bestseller” ribbon when you visit its Amazon webpage. That’s totally not shabby. I don’t think that very many books are #1 in any category after a year, and I hope it means that people think our book has lasting value.

Amazon also does provide actual sales numbers to authors, courtesy of Nielsen BookScan. Those numbers update once a week instead of once an hour. I know that people reading this blog have been curious about it in the past, so here is a graph of the first year of retail sales for The Book of Why, week by week. This isn’t actually all sales — for example, textbook and book club sales are not tabulated — but it’s close enough to give an author an idea of how the book is doing.

A few annotations:

1. New York Times bestseller territory is about 3000+ or maybe even 4000+ copies per week. As you can see, we never came within a country mile of that level of popularity. Some dreams are just plain unrealistic.
2. A few of the jumps have identifiable causes. The one-week jump in early June was most likely due to the New York Times review. The really dramatic jump in December was surely attributable to our selection as a Top Science Book of 2018 by Science Friday. Unlike the New York Times review, that honor seems to have boosted our sales for several weeks.
3. Also, the valley in October-November had an identifiable cause — our book was out of stock at Amazon due to supply chain issues.
4. For the last few weeks, sales have been pretty steady at 150 to 200 per week. The publisher (Basic Books) is very happy with this.

Who knows what the next year holds in store? For all the people who have sent their compliments, criticisms and questions to Judea and me (mostly to Judea), we have appreciated hearing from every one of you.

Coming to the U.K. on May 2!

Last week I got a package in the mail from England: ten copies of the paperback edition of The Book of Why! The occasion brought to mind that goofy old Beatles song, Paperback Writer. In fact, I’ve been waiting quite a long time to be a “paperback writer.”

My first book, The Big Splat, never sold enough to make it to paperback. My editor at Wiley told me that to be even considered for a paperback, it would have to sell 8000 copies. Ten years later it sputtered out and died at around 7500 copies sold. 

My second book, The Universe in Zero Words (now renamed The Story of Mathematics in 24 Equations) did make it to paperback, but under somewhat odd circumstances that make it hard for me to think of it as a true success. I’ll spare you the full details, but Elwin Street (the London publisher) took back the rights from Princeton University Press so that they could publish a paperback in the U.K. The book now has no U.S. publisher. It was more bad news than good news, in my opinion, because I really liked publishing with Princeton University Press.

And that brings us to The Book of Why, which has been all good news, all the time! Penguin UK, the British publisher, is releasing a paperback version on May 2. Note that this edition cannot be sold in the U.S. If you go to Amazon and try to buy it, and give them a U.S. address, I don’t think they will sell it to you.

So I am now going to be a paperback writer twice over in the U.K., but I am still not a paperback writer in the U.S.! According to our editor at Basic Books, the U.S. publisher of The Book of Why, their paperback version is scheduled for release in February 2020. I probably should not say too much about the reasons, but if any of you are wondering whether the sales are weak, the answer is no — in fact, it’s quite the opposite. The book continues to sell very well, and the editor is very happy with it. But as I said, the decision on when to go to paperback is a business decision, and the publishing business has changed in some obvious and less obvious ways. It’s no longer automatic that a successful book will come out in paperback after a year in hardback.

Which means, for my friends and readers in the U.S., if you’ve been saying, “I’ll wait for the paperback,” you probably shouldn’t. February 2020 is a long time away. I would recommend either buying the hardback, or taking a trip to England (a very good idea anyway!) and buying the Penguin paperback.

P.S. One good thing about buying the paperback is that you are getting the definitive, corrected version. If you are in the U.S. and you purchased the hardback, you should check the copyright page and see what the number of the printing is. If it is the fifth printing or later, it has been updated with very minor corrections and revisions. If you have a fourth printing or earlier, it is the original version, and we can send you a list of errata upon request.

Yesterday Judea Pearl, my co-author, sent me a link to a fascinating article that I consider to be even better than a book review. It is something that every author dreams of — positive evidence that our book is making a difference in the real world.

In the article, a group of 47 editors of research journals in the fields of respiratory, sleep, and critical care have compiled “best practices” recommendations for researchers who want to make causal inferences from observational data. That’s a mouthful, so let me explain. Our book, The Book of Why, is all about when you can and cannot answer causal queries on the basis of observational data (*). In short, we argue that if you can formulate an explicit causal model, represented in a simple dot-and-arrow diagram, then there are simple mathematical tests to determine what causal questions you can answer. Questions like “Does this drug affect the likelihood of heart attack?” or “Will working a midnight shift increase my risk of sleep disorders?” In other words, the kinds of questions that patients ask doctors all the time.

The official policy of many journals, including the Journal of the American Medical Association (stated in 2017 in an editor’s blog post) is that authors of medical articles are not allowed to answer such questions. “If it isn’t [a randomized trial], and is a report of an observational study… then all cause-and-effect language must be replaced.” (Emphasis added.)

So you can see the strength of the taboo that Judea and I are challenging. That is why the joint statement of 47 editors from a variety of medical journals is so important. It is a public admission that the guidance of the American Medical Association is outmoded and outdated. We do have the right to use cause-and-effect language, provided that we state our assumptions publicly and transparently.

It would be wrong, of course, for Judea and me to claim credit for this change in attitudes. It has been evolving for some years. However, I think it is very intriguing that Table 2 of the article lists The Book of Why at the top of the suggested resources for people wanting to learn more about causal inference. In fact, there are three books recommended: The Book of Why, Judea’s much more technical book Causality, and a forthcoming (2019) book by Jamie Robins and Miguel Hernan called Causal Inference.

Further, the article says,

We urge authors to consider using causal models when testing causal associations. The scientific, mathematical, and theoretical underpinnings of causal inference, developed by Judea Pearl, James Robins, Miguel Hernan, and others, have evolved sufficiently to permit the everyday use of causal models.

Allowing for the usual opacity of scientific writing, this is just about as ringing an endorsement of Judea’s methods as you could possibly ask for. I encourage anyone who is interested in doing causal analysis of observational studies to take a look at the full article, read our book, and connect with Judea on Twitter.

(*) Here “observational data” is intended to contrast with data gathered from a randomized controlled trial, which has long been the gold standard for establishing causal relationships. (In fact, knee-jerk acceptance of randomized controlled trials is almost as bad as knee-jerk rejection of observational studies. Both of them rely on assumptions that must be scrutinized. But that is a discussion for another time.)

This morning, my co-author Judea Pearl sent me an e-mail: “Did you know about this?” Well, no, I didn’t. Here’s what “this” means:

Science Friday, probably the nation’s most-listened-to radio show for science news, has named The Book of Why as one of its best science books of the year for 2018. It was one of eighteen books chosen. Here they are in alphabetical order. (That way, The Book of Why is near the top instead of at the bottom, the way it is on the Science Friday web page!) I’ll leave out the subtitles except when it’s necessary to understand what the book is about.

• Bad Blood: Secrets and Lies in a Silicon Valley Startup, by John Carreyrou
• The Book of Why, by Judea Pearl and Dana Mackenzie
• Broad Band: The Untold Story of the Women Who Made the Internet, by Claire Evans
• Chesapeake Requiem, by Earl Swift
• The Dialogues: Conversations about the Nature of the Universe, by Clifford Johnson
• The Dinosaur Artist, by Paige Williams
• Dispatches from Planet 3, by Marcia Bartusiak
• Heart: A History, by Sandeep Jauhar
• How to Change Your Mind, by Michael Pollan
• In Search of the Canary Tree, by Lauren Oakes
• Losing the Nobel Prize, by Brian Keating
• The Mystery of the Exploding Teeth, by Thomas Morris
• Nine Pints, by Rose George
• The Rise and Fall of the Dinosaurs, by Steve Brusatte
• Rising: Dispatches from the New American Shore, by Elizabeth Rush
• She Has Her Mother’s Laugh, by Carl Zimmer
• Spying on Whales, by Nick Pyenson
• What the Eyes Don’t See, by Mona Hanna-Attisha

Now I have to admit: I haven’t read any of these books, except for the one that I (co-)wrote. Shame on me! I do have In Search of the Canary Tree on my Amazon wish list, because I read a glowing review of it this week in Science. I am impressed that Science Friday says that She Has Her Mother’s Laugh is “clearly Zimmer’s best book,” because he has written a lot of books. Broad Band and Nine Pints (a book about blood) split my vote for the best title. In general, blood had a pretty good year (Nine Pints, Heart, and metaphorically speaking, Bad Blood). Of course, climate change had a good year; Chesapeake Requiem, Rising, and In Search of the Canary Tree all have to do with that topic. And dinosaurs are another evergreen topic (The Dinosaur Artist, The Rise and Fall of the Dinosaurs).

As for me, I’m just happy to be sitting at the same banquet table as these folks. It gives me the same spine-tingling, “Wow, is this happening?” feeling that I had the week our book came out. It’s possible that Judea and I will make it onto some other year-end best-of lists (I certainly hope so!), but this is the first one, so it’s special to me. And it’s a pretty great list to be on.

If any of you have visited my book’s page on Amazon recently, you have seen a scary message like this one: “Available only through third-party vendors.” I don’t see that message today, but the page still does not say “In Stock,” nor does it give any hint of how long it will take you to get a copy. (Last week it said “Usually ships in 2 to 4 weeks.”)

What is the problem? The short answer: Amazon is sold out. Yes, that’s right, the company that has everything does not have a single copy of The Book of Why.

But there’s a more interesting answer behind that. The publishing industry has (in the words of my editor) been in a state of utter bedlam this fall because of two things. First, we are experiencing one of the worst paper shortages in years. (Bet you didn’t know that.) Second, there is an apparent undersupply of labor and printing press availability. According to this article in Publishers Weekly, the 5th largest printer in the U.S. unexpectedly went out of business this year, and the 3rd largest printer is in bankruptcy. (It’s expected to survive.)

Even for publishers who don’t use those printers, there is still a ripple effect as everybody scrambles to find a printer with available capacity. My editor tells me that they ordered the next printing of my book (the fifth, I think, or maybe the sixth), 4000 copies, back in early September. Ordinarily, he said, that order would take 15 days to fill. But they’re still waiting. A few days ago he told me that the books were expected in the warehouse on November 2, but now he’s telling me that they will come on November 16. Which means that Amazon will not have any copies of The Book of Why until then, unless they go looking under park benches and in dumpsters.

There are a couple of things to be thankful for, although only in a silver-lining-dark-cloud sense. First, The Book of Why is still available in bookstores and at Barnes and Noble online. Perhaps this printing shortage will weaken Amazon’s grip on the U.S. book market, which is not necessarily a bad thing. Second, as my editor points out, at least The Book of Why came out last spring, and so it had about five months of relative normalcy before the printing shortage began to affect sales. That means it reached a lot of people already, it developed good word-of-mouth, and I don’t think that the supply shortage will affect it in the long run. But for new books that came out in October, the shortage could be a kiss of death, preventing them from ever gaining momentum and generating word of mouth sales.

You can really see the effect of the printing shortage in my book’s latest Nielsen BookScan totals. Here is a graph of its weekly sales for the first 24 weeks of its existence:

For the first 23 weeks, it looks like a Poisson distribution with added noise, and perhaps a heavier tail. For non-mathie folks, here is what Poisson distributions look like. I think that the green one is the best match for the Book of Why sales graph.

But then in week 24, the sales dropped from 239 books to 81 — nearly a threefold drop! I expect that week 25 will be even worse, with no Amazon sales. Amazon is supposedly about two-thirds of the U.S. book market.

(By the way, you can see a few other interesting things about the Book of Why sales graph. Our peak, in week 2, was 1459 copies. As a rough estimate, you need to sell about 3000 copies a week to make it to the hardcover bestseller lists. So after I saw the downturn in week 3, I knew we were never going to be an official bestseller. The second peak, 1254 copies in week 4, came after the New York Times reviewed The Book of Why. The third peak, 567 copies in week 9, came for unknown reasons. Maybe Amazon just lowered its price that week. The fourth peak, 532 copies in week 14, came after our book was made an alternate selection for something called the Next Big Idea Club. I am still waiting for the fifth peak… maybe after the supply crisis abates.)

All this stuff about printers reminded me of something I almost never think about any more. When I was in college, I worked for four summers at a printing press. Although I was in the front office and didn’t go back into the production area too often, I always enjoyed it when I got the chance. It was a technological marvel. Those huge machines slap ink onto paper and fold it up and bind it faster than the eye can see.

That experience taught me that a book or a magazine is not just a collection of fine words and brilliant ideas. It is, first and foremost, a physical object that somebody has to produce. Those beautiful ideas would be worth nothing if we didn’t have the physical means to manufacture and distribute books in large quantities. The Internet has changed things a little bit, but not as much as you’d think. Gutenberg’s invention of the printing press (and all the improvements that have followed) is still king. You just don’t appreciate it until suddenly you don’t have it any more.

Once upon a time, I wrote a book. Interestingly, the idea for the book was not mine. A book packager named Elwin Street Productions, in London, came to me with a proposal for a history of the greatest equations in mathematics. The book came out in 2012 as The Universe in Zero Words, co-published by Elwin Street and by Princeton University Press.

To understand this story fully, you do need to understand a little bit about the book business. A “packager” is not a publisher per se. What they do is put together a book package — words by an author, images from various sources — that they shop around to traditional publishers. It’s a somewhat common procedure for image-intensive books. (Think Time-Life books, for example.) For me, it was fun to work on a book where the images would be such a big part of the finished product. This went along very well with a constant theme of the book, the idea that mathematical equations can be beautiful.

Over the years the book was translated into other languages, with other titles. The French edition was Fous d’Equations (Crazy about Equations), which I thought was a great title. The Italian edition was more or less a direct translation, L’Universo senza parole (The Universe without Words) but had a different subtitle (Mathematics Revealed in 24 Equations). 

Everything was fine until last winter, when I was informed, to my surprise and dismay, that Elwin Street was leaving the co-publishing agreement with Princeton University Press. In part, the reason is that they are converting their business so that they will be an actual publishing house. Therefore, they wanted to be able to publish my book themselves in the U.K. market. I do not know or understand why they wouldn’t let PUP continue to publish the book in the U.S. I definitely do not agree with their decision. However, because Elwin Street owns all the rights, it’s their decision to make. They are currently looking for a new U.S. publisher.

Anyway, for every dark cloud there is a silver lining. My book now has a new edition in England, and it has yet another new title! It’s now called The Story of Mathematics in 24 Equations, which I think is probably more appealing and less confusing than the original title. I now think that the original title, The Universe in Zero Words, even though it was my idea, was not a particularly good one. The only people who understand it right away are mathematicians; for everyone else, I have to explain it. 

For any readers in England who might come across it in a bookstore, the new edition is almost identical to the old one. The figures are the same. I made a few small corrections to the text, but there have been no major additions or deletions. The main differences are that it is a paperback and it has a beautifully re-designed cover, which goes better with the design scheme of the interior. It remains a gorgeous book to look at, thanks to all the color pictures. If you purchased the old edition, there really isn’t any need to buy the new one. However, I suspect that there is a large market out there of people who never heard of the first edition, and I am delighted that Elwin Street is giving my old book a second chance to reach them.

I hope that American readers will also get another chance (eventually) to buy The Book of Many Titles, but we’ll have to wait and see!

Today Jen Golbeck’s interview with Judea and me for the Science (magazine) Podcast went up, and you can listen to it here. (Alternatively, for the next 7 days, you can also go to the main Science magazine website at www.sciencemag.org and scroll down to the podcast.) The book review segment starts at 18:48 with a long-ish introduction by Golbeck, and the actual interview starts at 20:35 and lasts to 26:05.

Some thoughts: I do wish we had had more time. Basically Judea and I got to answer two questions each, and we got to discuss only about three topics out of a 400-page book. I have to blame myself for giving too rambling an answer to my first question. I kept saying “and this…” “and that…” If you wrote a transcript of the interview, you would probably find that I crammed about five sentences into one. A good thing to avoid in the future!

On the positive side, I thought Judea sounded very patient, reasonable, and persuasive. I think that between the two of us, we very clearly expressed one of our central ideas, which is that causal diagrams allow scientists to express their assumptions in such a way that they are visible and can be discussed by others. A point that was less apparent is that causal diagrams also guide you toward the right way to analyze your data.

One interesting point for very sharp-eared listeners: this may well be the first Science Podcast with a laugh track! My wife was listening to the interview as we recorded it, and at the 25:41 mark I said something that made her laugh. I think that Jen Golbeck tried to edit it out, but didn’t completely succeed. (It was when I talked about an “outbreak of common sense.”) 

Outtakes: What Didn’t Make it into the Podcast?

Golbeck asked each of us, “If you could put your book on the desk of one person, who would it be?” Judea said he would like the book to be read by economists. He feels that economics is one discipline that has resisted his graphical approach to causality.

My answer was “Ruth Bader Ginsburg.” In our book we discuss the long tradition of counterfactuals in the law. (A defendant is guilty if the victim would have been unharmed but for the defendant’s actions. “Would have” and “but for” express a counterfactual condition.) I would like lawyers to be aware that some scientists now are taking counterfactuals seriously, and that we have algorithms that can perhaps improve our ability to make such judgements. In particular, I know that some members of the Supreme Court are skeptical of gerrymandering measures that rest on counterfactual conditions, and I would like to alleviate some of their concern.

Another answer that was edited out: I told Golbeck about a question that Judea once asked me, about a month after the book came out: “Why was it necessary for me to go behind my colleagues’ back and address the public in order to get my colleagues’ attention?” My answer was: “First, I’m not sure you did go behind their backs. But if you did, it’s only because they themselves turned their back on the public.” I think that this experience has totally sold Judea on the fact that it’s not just good publicity to face the public — it’s good science, too.

P.S. It was a great day for Santa Cruz on the Science Podcast! The interview just before ours was with Emily Brodsky of UC Santa Cruz, who talked about human-induced earthquakes.

The American Association for the Advancement of Science, publishers of Science magazine, also have a weekly podcast, called (of course) the Science Podcast. I’m pleased to announce that The Book of Why will be the featured book two weeks from now, in the last week of August.

This makes up for my (mild) disappointment that Science never published a review of The Book of Why in the print magazine. Of course, there are many science books published every week, and it’s not easy to write one that stands out enough to merit a review. The way I look at it, a review on the podcast is even more special than a review in print, because only one book a month gets chosen for the podcast, as opposed to three or four a week that get reviewed in the magazine.

Yesterday Jen Golbeck, the host of the book review segment, interviewed Judea and me by telephone (we had a conference call). I thought it went really well and she asked some great questions. The one I liked best was, “If you could put your book on the desk of one person, who would it be?” I hope you’ll get to hear my surprising answer. I can’t guarantee that it will make the episode, because the interview lasted 28 minutes but it will have to be edited down to 7 or 8 minutes.

My wife listened in on the interview, and laughed at a couple of the things Judea said. He really has a wonderful, understated comic wit. I imagine the chuckles will be edited out, but it would be funny if they said, “This episode was recorded in front of a live audience of one person.” 

Even without the laugh track, I’m sure that the interview will be interesting to listen to! After it goes live I’ll post a link to it here.

Earlier this week I gave a presentation based on The Book of Why at the Simons Institute for the Theory of Computing at UC Berkeley, where I have been a journalist in residence for the last eight weeks. It was very similar to my presentation at the Aptos Library in Santa Cruz, with one important difference: this one was video recorded! You can watch the whole talk at this link on YouTube, or you can watch it on the embedded viewer below. In two days, it has already gotten more than 400 views!

Just one word of warning: The first three minutes of the video is devoid of content; we were just chatting and waiting for latecomers to settle down. Beginning around the 3:00 mark, Shafi Goldwasser introduces me. If you want to skip the introduction and just listen to the actual lecture, it starts around 4:30.

Some good questions were asked during the lecture. I’d like to say a little bit more about two of them. Around the 39:30 mark, one listener asks whether an inference of causation might require a higher level of significance (such as p < 0.01) in the presence of a confounder. The answer I gave was correct but I could have explained it a bit better. I should have reminded him of the earlier example of the walking study (from about 23:00 to 23:20). The association between walking and reduced mortality was very strong and would probably have passed the test he suggested. Nevertheless, the association would be completely spurious if, say, the people in the intense-walking group were 10 years younger than the casual walkers. In that case the association would be completely explained by the age difference. So a low p-value, in the presence of a confounder, does not give us any guarantees about causality.

Secondly, one listener said something that I unfortunately talked over, and it’s hard to make out what he said. At 52:30, he suggested that one way a researcher could present a causal finding would be this: “Either you believe that smoking causes lung cancer, or there exists a confounding factor. Given our experience, we do not believe there is such a confounder.” I like this way of putting it because it puts the responsibility on a skeptic: if you don’t believe in this result, you have to say what specifically is wrong about our causal diagram. Perhaps we have omitted a confounder, or assumed two variables do not have a direct causal relation when in fact they do. In this way, we direct the discussion precisely where it should be directed: What do we know about the web of causes and effects, what is it reasonable to assume, and what do we not know yet? If we disagree about some aspect of the diagram, how sensitive are the results to our choice?

Not all of the lecture is quite this serious and academic! Keep an eye out for cute puppy pictures at 14:39 and a cute bunny picture at 50:13!