Skip to content

Erdos bio for kids

Chris Gittins recommends the book, “The Boy Who Loved Math: The Improbable Life of Paul Erdos,” by Deborah Heiligman. Gittins reports:

We read it with our soon-to-be-first-grader this evening. She liked it and so did we. I knew a little about Erdos but the book probably quadrupled my knowledge. Thought it might be of interest to readers of your blog who have little kids – and perhaps even to those who don’t.

I haven’t read any books about Erdos myself, as I’ve always been a bit creeped out by the Erdos thing, maybe because I don’t like how it reinforces the idea of mathematicians as weirdos. But I’ll pass along the recommendation. A book called “The Boy Who Loved Math”—I would’ve loved that myself as a kid.


  1. adam says:

    With the ever-rising number of amphetamine prescriptions for children I wonder if there is a chapter that describes Erdos’ fondness for amphetamines as a relatable moment.

  2. Steve Laniel says:

    That children’s book title is a significantly softened version of the adult book title: “The Man Who Loved Only Numbers”.

  3. Richard McElreath says:

    This book was a hit with my son. We read it every night for a week, when we first got it.

  4. Rahul says:

    Ha, you sound rather meh about two of the most fascinating scientists (I think): Feynman & Erdos.

  5. Sean says:

    I’ve gotta agree with the weirdo thing.

  6. Dzhaughn says:

    Andrew, the only reason you are creeped out by this sort of thing is because you are a numbers person, not a people person. :)

  7. Martha says:

    “I’ve always been a bit creeped out by the Erdos thing, maybe because I don’t like how it reinforces the idea of mathematicians as weirdos.”

    I’ve thought of Erdos as eccentric, but never as a weirdo. (I checked this out with a colleague, who agrees.) But possibly there is a generational thing going on here — I think I first encountered Erdos while I was an undergraduate, in the early 60’s (before Andrew was born?). I don’t recall whether he was asked to speak in one of my classes, or whether he gave a special talk for undergraduates, or whether the honors math students were encouraged to attend the talk he was giving for the department.

    I recall encountering him on at least a couple of other occasions thereafter. One I remember the details of; the other(s) I’m fuzzy on as far as location. Some things I remember from hearing him talk (either in formal talks or in conversations at parties) include: Calling young people “epsilons,” his offering prizes for solutions of problems, his talking about “The Book.” I remember his once saying, “I know this is true, but I haven’t proved it yet.” I also learned from others about his peripatetic lifestyle, how Ron Graham acted as his banker, and, of course, Erdos numbers (Mine is 4, which can be traced by at least 3 paths.)

    For me, he’s an integral part of the mathematical landscape.

    • Andrew says:


      My reaction to Erdos is similar (although not identical) to my reactions to Neumann and Feynman, two other great mathematicians/physicists. Everyone seemed to love these guys, but when I read about them, they just seem so full of themselves and obnoxious. But . . . everyone seemed to love these guys, so they probably really were lovable in real life. There’s just something about the stories people tell about them, that turn me off. In contrast, when I read about Stanislaw Ulam, I see that lots of people didn’t like him so much, but I get the impression that I would’ve liked him a lot.

      Just as a point of reference: I met John Conway once and liked him. I didn’t meet him in any formal sense, indeed I wasn’t even introduced to him. I just happened to be at a conference at Princeton once and wandered into the math lounge, I met this guy who showed me a really cool wooden puzzle (so cool that I went home and spent several hours with the saw and sander and constructed one myself). It was only later that I realized this must have been the famous Mr. Life. He wasn’t “charming” but I liked him. He seemed like my kind of guy.

      I also met Freeman Dyson once, but this was for about 30 seconds at some conference, so I just said hi and that was about it.

      • Rahul says:

        Feynman was no-nonsense. He asked blunt, practical questions. It’s like asking after a conference talk why the work presented was useful. It tends to upset people. And label you as obnoxious.

        I rather wish we had a lot more of these “obnoxious” geniuses. Sending Feynman to a modern Psych. Seminar would probably do them a lot of good.

        • Andrew says:


          That aspect of Feynman didn’t bother me. Actually I saw Feynman give a talk once and it was an excellent talk, he was very open about what he knew and where he was speculating. I wasn’t so thrilled with how Feynman told those obnoxious stories. Phil and I use the term “Feynman story” for any anecdote that someone tells that is structured so that the teller comes off as a genius and everyone else in the story comes off as an idiot.

          • Martin says:

            I admit that this always bothered me, too, w/r/t Feynman. It must be hard not to tell stories about how one was by far the smartest guy in the room if one usually was, by far, the smartest guy in the room. Paradoxical though as it might seem, this is also why it is all the more disappointing that he didn’t manage to frame his anecdotes differently. And at least to a certain extent, I fancy that Feynman’s Camel ad-like (self-)image explains more about his popularity than his fans would like to admit.

    • jrc says:

      From someone who never even came close to meeting him, my impression is also closer to eccentric than to weirdo. And I kinda like the idea of embracing our eccentrics. For that matter, our weirdos too.

      Andrew – is it just that you think the idea of “mathematician as weirdo” is discouraging to students interested in math or reinforces stereotypes that lead to bullying? Because from a broader perspective, the idea of exposing kids to fantastic people who are also well outside the mainstream seems pretty OK to me.

      • I think holding up eccentrics like Erdos or Godel, or more recently Grigori Perelman who famously turned down a couple million bucks for proving something that implied the Poincare Conjecture gives people the impression that they can’t do important things in math or other technical fields unless they are eccentric.

        That’s just bullshit, but it is definitely the “vibe” you get from a steady stream of books about eccentrics who do great things.

        • Rahul says:

          I think you overestimate people’s naivete. I doubt we are losing any significant crop of top notch mathematicians just because they don’t find themselves weird enough.

          I don’t think anyone’s intentionally glamorizing the eccentricity pf Erdos or Perelman. They are what they are. To paint Perelman as the typical guy next door just wouldn’t be factual. And the correlation between eccentricity and genius seems real enough though obviously eccentricity is neither necessary nor sufficient.

          Then again, for even Perelman there’s a Tao. The latter seems pretty well adjusted. The world is what it is.

  8. Rahul says:


    Maybe. But personally, I think we have too few surgeons and too many mathematicians anyways. So I’m hardly sad if wannabe middling mathematicians end up being middling surgeons.

    In any case, I think a mediocre surgeon can have a lot more positive social impact than a mediocre mathematician. Perhaps I’m wrong.

    • Bill Jefferys says:

      It all depends on how mediocre the surgeon is. A mediocre mathematician is unlikely to kill a student in classroom instruction; a mediocre surgeon may kill a patient on the operating table who would not have been killed had the surgeon been more skilled.

      • Andrew says:


        A bunch of mediocre mathematicians did their best to scuttle my career a few years ago. On the other hand, (1) they didn’t succeed, and (2) it wouldn’t’ve been as bad as killing a patient on an operating table.

      • Rahul says:

        Is it safe to assume that a 70 percentile ability, US-trained surgeon is not at a high risk of killing patients to the point where we’d rather not want him around?

        • Andrew says:

          As the character in Straight Man would say, Orshee.

        • Bill Jefferys says:

          The point of my comment is that in decision making, the loss function is as important as if not more important than the probability calculation.

          If I were going to be operated on by a surgeon, I would want to know that surgeon’s track record. Number of surgeries of that particular kind successfully conducted, number that failed, number that resulted in death, etc. I wouldn’t have much choice if I had been rushed to the ER, but if I had an “elective” situation I would (and have) checked carefully.

        • Bill Jefferys says:

          One would hope that a 1% ability, US-trained surgeon would not be at high risk of killing patients to the point where we’d rather not want hir around.

          But without statistics on actual surgeons’ performance in the death department, I cannot say whether the correct number in fact is 1% or 70%. Can you?

          I suspect that this number was picked out of a certain nether region.

    • There’s a lot more to using mathematical ability than “being a mathematician”. I’d argue that the logical process in mathematics would make you a much much better say internist or infectious disease specialist, or epidemiologist than it would surgeon. Going ahead with mathematical training and learning techniques of proof, statistical reasoning, etc is going to make you a better doctor when the kind of doctor you are is one that needs to do diagnosis, decision making under uncertainty, tradeoffs between disease risk and treatment risk, and soforth all day long.

      My impression is that the rules of thumb used by doctors to “rule out” things works great when you’re in the appropriate category, and leads to misery when you get improperly categorized and the doctors refuse to go down the proper path because something that is true in 90% of patients is treated as if it’s true in 100% of patients, and so you can’t have disease X.

      In my experience, the thing keeping the supply of doctors down is other doctors, especially those in charge of residency programs and the board exams etc. Don’t blame any perceived lack of supply of doctors on mathematicians!

      • Martha says:

        “My impression is that the rules of thumb used by doctors to “rule out” things works great when you’re in the appropriate category, and leads to misery when you get improperly categorized and the doctors refuse to go down the proper path because something that is true in 90% of patients is treated as if it’s true in 100% of patients, and so you can’t have disease X.”

        Or they never think of the true problem (their training may never have mentioned it) and keep looking for the problems that cause the symptoms in most patients. An example I’ve experienced: Severe gastric distress isn’t necessarily caused by a direct gastric problem — it can be the result of a pulled abdominal muscle. (None of primary care physician, anesthesiologist, nor gastroenterologist considered this possibility, even though the onset was shortly after something seemed to snap while doing abdominal exercises. But when I finally suggested a physical therapist, my PCP said it was worth a try. Guess what — the physical therapist diagnosed the problem, and I could eat half of a normal meal after one PT session, for the first time in several weeks. But I’ve never regained full use of the muscle for things like distance walking, and keep wondering if it would have healed better if I had gotten treatment earlier.)

        • Martha, yes exactly, obviously doctors can’t know everything, so they specialize in a certain area, but I think the mathematical/logical ability to reason through something and figure things out, to take shortcuts but know that they are shortcuts, to backtrack, recalculate, weigh risks vs benefits quantitatively (instead of the usual way which is more of a “gut instinct” method) etc etc. These are mathematical skills doctors should have.

          All this is to say, holding up Erdos or Perelman as the epitome of mathematics in the popular press/books I think discourages people from developing mathematical skills that can benefit them a lot even as non-mathematicians.

          I can not tell you how often doctors get diagnostic test X when the outcome isn’t really going to change their course of action, or FAIL to get diagnostic test Y because they’ve already made up their mind prematurely but it’s pretty often based on my similar kinds of experiences.

          • For example, a doctor once listened to my lungs and said “well, you’ve earned yourself a chest X-ray”. I responded “is it going to change your treatment choice? Because maybe we could just do the treatment and only do the X-ray afterwards if the treatment doesn’t work”. She admitted that this is actually what we should do, but that most patients actually like getting chest X-rays it makes them feel better that something is getting done…. sigh

            • Martha says:

              Is it that “most patients actually like getting chest X-rays it makes them feel better that something is getting done” or that doctors believe that “most patients actually like …”? I can’t help but wonder if it’s the latter.

  9. cheese_d says:

    It sounds like an interesting book, though by the title, I was thinking it would be a book about Galois.

Leave a Reply