Lessons from Bowen and Darryl

At the JMM this year, I had the pleasure of attending a minicourse on “Designing and Implementing a Problem-Based Mathematics Course” taught by Bowen Kerins and Darryl Yong, the masterminds behind the legendary PCMI teachers’ program Developing Mathematics course, with a significant assist from Mary Pilgrim of Colorado State University.

I’ve been wanting to get a live taste of Bowen and Darryl’s work since at least 2010, when Jesse Johnson, Sam Shah, and Kate Nowak all came back from PCMI saying things like “that was the best math learning experience I’ve ever had,” and I started to have a look at those gorgeous problem sets. It was clear to me that they had done a lot of deep thinking about many of the central concerns of my own teaching. How to empower learners to get somewhere powerful and prespecified without cognitive theft. How to construct a learning experience that encourages learners to savor, to delectate. That simultaneously attends lovingly to the most and least empowered students in the room. &c.

I want to record here some new ideas I learned from Bowen and Darryl’s workshop. This is not exhaustive but I wanted to record them both for my own benefit and in the hopes that they’ll be useful to others. In the interest of keeping it short, I won’t talk about things I already knew about (such as their Important Stuff / Interesting Stuff / Tough Stuff distinction) even though they are awesome, and I’ll keep my own thoughts to a minimum. Here’s what I’ve got for you today:

1) The biggest takeaway for me was how exceedingly careful they are with people talking to the whole room. First of all, in classes that are 2 hours a day, full group discussions are always 10 minutes or less. Secondly, when students are talking to the room it is always students that Bowen and Darryl have preselected to present a specific idea they have already thought about. They never ask for hands, and they never cold-call. This means they already know more or less what the students are going to say. Thirdly, they have a distinction between students who try to burn through the work (“speed demons”) and students who work slowly enough to receive the gifts each question has to offer (“katamari,” because they pick things up as they roll along) – and the students who are asked to present an idea to the class are only katamari! Fourthly, a group discussion is only ever about a problem that everybody has already had a chance to think about – and even then, never about a problem for which everybody has come to the same conclusion the same way. Fifthly, in terms of selecting which ideas to have students present to the class, they concentrate on ideas that are nonstandard, or particularly visual, or both (rather than standard and/or algebraic).

This is for a number of reasons. First of all, the PCMI Developing Mathematics course has something like 70 participants. So part of it is the logistics of teaching such a large course. You lose control of the direction of ideas in the class very quickly if you let people start talking and don’t already know what they’re going to say. (Bowen: “you let them start just saying what’s on their mind, you die.”) But there are several other reasons as well, stemming (as I understood it anyway) from two fundamental questions: (a) for the people in the room who are listening, what purpose is being served / how well are their time and attention being used? and (b) what will the effect of listening to [whoever is addressing the room] be on participants’ sense of inclusion vs. exclusion, empowerment vs. disempowerment? Bowen and Darryl want somebody listening to a presentation to be able to engage it fluently (so it has to be about something they’ve already thought about) and to get something worthwhile out of it (so it can’t be about a problem everybody did the same way). And they want everybody listening to feel part of it, invited in, not excluded – which means that you can’t give anybody an opportunity to be too high-powered in front of everybody. (Bowen: “The students who want to share their super-powerful ideas need a place in the course to do that. We’ve found it’s best to have them do that individually, to you, when no one else can hear.”)

2) Closely related. Bowen talked at great length about the danger of people hearing somebody else say something they don’t understand or haven’t heard of and thinking, “I guess I can’t fully participate because I don’t know that idea or can’t follow that person.” It was clear that every aspect of the class was designed with this in mind. The control they exercise over what gets said to the whole room is one aspect of this. Another is the norm-setting they do. (Have a look at page 1 of this problem set for a sense of these norms.) Another is the way they structure the groups. (Never have a group that’s predominantly speed-demons with one or two katamari. If you have more speed-demons than katamari, you need some groups to be 100% speed demon.)

While this concern resonates with me (and I’m sure everybody who’s ever taught, esp. a highly heterogeneous group), I had not named it before, and I think I want to follow Bowen and Darryl’s lead in incorporating it more essentially into planning. In the past, I think my inclination has been to intervene after the fact when somebody says something that I think will make other people feel shut out of the knowledge. (“So-and-so is talking about such-and-such but you don’t need to know what they’re talking about in order to think about this.”) But then I’m only addressing the most obvious / loud instances of this dynamic, and even then, only once much of the damage has already been done. The point is that the damage is usually exceedingly quiet – only in the mind of somebody disempowering him or herself. You can’t count on yourself to spot this, you have to plan prophylactically.

3) Designing the problem sets specifically with groupwork in mind, Bowen and Darryl look for problems that encourage productive collaboration. For example, problems that are arduous to do by yourself but interesting to collaborate on. Or, problems that literally require collaboration in order to complete (such as the classic one of having students attempt to create fake coin-flip data, then generate real data, trade, and try to guess other students’ real vs. fake data).

4) And maybe my single favorite idea from the presentation was this: “If a student has a cool idea that you would like to have them present, consider instead incorporating that idea into the next day’s problem set.” I asked for an example, and Bowen mentioned the classic about summing the numbers from 1 to n. Many students solved the problem using the Gauss trick, but some students solved the problem with a more visual approach. Bowen and Darryl wanted everybody to see this and to have an opportunity to connect it to their own solution, but rather than have anybody present, they put a problem on the next day’s problem set asking for the area of a staircase diagram, using some of the same numbers that had been asked about the day before in the more traditional 1 + … + n form.

I hope some of these ideas are useful to you. I’d love to muse on how I might make use of them but I’m making myself stop. Discussion more than welcome in the comments though.


Hard Problems and Hints

I have a friend O with a very mathematically engaged son J, who semi-often corresponds with me about his and J’s mathematical experiences together. We had a recent exchange and what I was saying to him I found myself wanting to say to everybody. So, without further ado, here is his email and my reply (my take on Aunt Pythia) –

Dear Ben,

J’s class is learning about volume in math. They’ll be working with cubes, rectangular prisms and possibly cylinders, but that’s all. He asked his teacher if he could work on a “challenge” that has been on his mind, which is to find a formula for the volume of one of his favorite shapes, the dodecahedron. He build a few of these out of paper earlier in the year and really was/is fascinated with them. I think he began this quest to find the volume thinking that it would be pretty much impossible, but he has stuck with it for almost a week now. I am pleased to see that he’s not only sticking with it, but also that he has made a few pretty interesting observations along the way, including coming up with an approach to solving it that involves, as he put it, “breaking it up into equal pieces of some simpler shape and then putting them together.” After trying a few ways to break/slice up the dodecahedron and finding that none of them seemed to make matters simpler, he had an “ah ha” moment in the car and decided that the way to do would be to break it up into 12 “pentagonal pyramids” (that’s what he calls them) that fit together, meeting at the center of rotation of the whole shape. If we can find the volume of one of those things, we’re all set. A few days later, he told me that he realized that “not every pentagonal pyramid could combine to make a dodecahedron” so maybe there was something special about the ones that do, i.e., maybe there is a special relationship between the length of the side of the pentagon and the length of the edge of the pyramid that could be used to form a dodecahedron.

He is still sticking with it, and seems to be having a grand time, so I am definitely going to encourage him and puzzle through it with him if he wants.

But here’s my question for you…

I sneaked a peak on google to see what the formula actually is, and found (as you might know) that it’s pretty complicated. The formula for the volume of the pentagonal pyramid involves \tan 54 (or something horrible like that) and the formula for the volume of a dodecahedron involves 15 + 7\sqrt{5} or something evil like that. In short, I am doubtful that he will actually be able to solve this problem he’s puzzling through. What does a good teacher do in such a situation? You have a student who is really interested in this problem, but you know that it’s far more likely that he will hit a wall (or many walls) that he really doesn’t have the tools to work through. On the other hand, you really want him to find satisfaction in the process and not measure the joy or the value of the process by whether he ultimately solves it.

I certainly don’t care whether he solves it or not. But I want to help him get value out of hitting the wall. How do you strike a balance so that the challenge is the right level of frustrating? When is it good to “give a hint” (you’ve done that for me a few times in what felt like a good way… not too much, but just enough so that the task was possible).

In this case, he’s at least trying to answer a question that has an answer. I suppose you could find a student working on a problem that you know has NO known answer, or that has been proven to be unsolvable. Although there, at least, after the student throws up his hands after giving it a good go, you can comfort her by saying, “guess what… you’re in good company!” But here, I’d like to help give him some of the tools he might use to actually make some headway, without giving away the store.

I think he’s off to a really good start — learning a lot along the way – getting a lot of out the process, the approach. I can already tell that many of the “ah ha” moments have applicability in all sorts of problems, so that’s wonderful.

Best, O

Dear O,

Wow, okay first of all, I love that you asked me this and it makes me really appreciate your role in this journey J is on, in other words I wish every child had an adult present in their mathematical journey who recognizes the value in their self-driven exploration and is interested in being the guardian of the child’s understanding of that value.

Second: no matter what happens, you have access to the “guess what… you’re in good company” response, because the experience of hitting walls as you try to find your way through the maze of the truth is literally the experience of all research mathematicians, nearly all of the time. If by any chance J ends up being a research mathematician, he will spend literally 99% or more of his working life in this state.

In fact, I would want to tweak the message a bit; I find the “guess what… you’re in good company” a tad consolation-prize-y (as also expressed by the fact that you described it as a “comfort”). It implies that there was an underlying defeat whose pain this message is designed to ameliorate. I want to encourage you and J both to see this situation as one in which a defeat is not even possible, because the goal is to deepen understanding, and that is definitely happening, regardless of the outcome. The specific question (“what’s the volume of a dodecahedron?”) is a tool that’s being used to give the mind focus and drive in exploring the jungle of mathematical reality, but the real value is the journey, not the answer to the question. The question is just a tool to help the mind focus.

In fairness, questing for a goal such as finding the answer to a question and then not meeting the goal is always a little disappointing, and I’m not trying to act like that disappointment can be escaped through some sort of mental jiu-jitsu. What I am trying to say is that it is possible to experience this disappointment as superficial, because the goal-quest is an exciting and focusing activity that expresses your curiosity, but the goal is not the container of the quest’s value.

So, that’s what you tell the kid. Way before they hit any walls. More than that, that’s how you should see it, and encourage them to see it that way by modeling.

Third. A hard thing about being in J’s position in life (speaking from experience) is that the excitement generated in adults by his mathematical interests and corresponding “advancement” is exciting and heady, but can have the negative impact of encouraging him to see the value of what he’s doing in terms of it making him awesome rather than the exploration itself being the awesome thing, and this puts him in the position where it is possible for an unsuccessful mathematical expedition to be very ego-challenging. This is something that’s been behind a lot of the conversations we’ve had, but I want to highlight it here, to connect the dots in the concrete situation we’re discussing. To the extent that there are adults invested in J’s mathematical precociousness per se, and to the extent that J may experience an unsuccessful quest as a major defeat, these two things are connected.

Fourth, to respond to your request for concrete advice regarding when it is a good idea to give a hint. Well, there is an art to this, but here are some basic principles:

* Hints that are minimally obtrusive allow the learner to preserve their sense of ownership over the final result. The big dangers with a hint are (a) that you steal the opportunity to learn by removing a part of the task that would have been important to the learning experience, and (b) that you steal the experience of success because the learner doesn’t feel like they really did it. These dangers are related but distinct.

* How do you give a minimally obtrusive hint?

(a) Hints that direct the learner’s attention to a potentially fruitful avenue of thought are superior to hints that are designed to give the learner a new tool.

(b) Hints that are designed to facilitate movement in the direction of thought the learner already has going on are generally better than hints that attempt to steer the learner in a completely new direction.

* If the learner does need a new tool, this should be addressed explicitly. It’s kind of disingenuous to think of it as a “hint” – looking up “hint” in the dictionary just now, I’m seeing words like “indirect / suggestion / covert indication”. If the learner is missing a key tool, they need something direct. The best scenario is if they can actually ask for what they need:

Learner: If I only had a way to find the length of this side using this angle…
Teacher: oh yes, there’s a whole body of techniques for that, it’s called trigonometry.

This is rare but that’s okay because it’s not necessary. If the teacher sees that the learner is up against the lack of a certain tool, they can also elicit the need for it from the learner:

Teacher: It seems like you’re stuck because you know this angle but you don’t know this side.
Learner: Yeah.
Teacher: What if I told you there was a whole body of techniques for that?

Okay, those are my four cents. Keep me posted on this journey, it sounds like a really rich learning experience for J.

All the best, Ben

A Critical Language for Problem Design

I am at the Joint Mathematics Meetings this week. I had a conversation yesterday, with Cody L. Patterson, Yvonne Lai, and Aaron Hill, that was very exciting to me. Cody was proposing the development of what he called a “critical language of task design.”

This is an awesome idea.

But first, what does he mean?

He means giving (frankly, catchy) names to important attributes, types, and design principles, of mathematical tasks. I can best elucidate by example. Here are two words that Cody has coined in this connection, along with his definitions and illustrative examples.

Jamming – transitive verb. Posing a mathematical task in which the underlying concepts are essential, but the procedure cannot be used (e.g., due to insufficient information).

Example: you are teaching calculus. Your students have gotten good at differentiating polynomials using the power rule, but you have a sinking suspicion they have forgotten what the derivative is even really about. You give them a table like this

x f(x)
4 16
4.01 16.240901
4.1 18.491

and then ask for a reasonable estimate of f'(4). You are jamming the power rule because you’re giving them a problem that aims at the concept underlying the derivative and that cannot be solved with the power rule.

Thwarting – transitive verb. Posing a mathematical task in which mindless execution of the procedure is possible but likely to lead to a wrong answer.

Example: you are teaching area of simple plane figures. Your students have gotten good at area of parallelogram = base * height but you feel like they’re just going through the motions. You give them this parallelogram:
Of course they all try to find the area by 9\times 41. You are thwarting the thoughtless use of base * height because it gets the wrong answer in this case.

Why am I so into this? These are just two words, naming things that all teachers have probably done in some form or another without their ever having been named. They describe only a very tiny fraction of good tasks. What’s the big deal?

It’s that these words are a tiny beginning. We’re talking about a whole language of task design. I’m imagining having a conversation with a fellow educator, and having access to hundreds of different pedagogically powerful ideas like these, neatly packaged in catchy usable words. “I see you’re thwarting the quadratic formula pretty hard here, so I’m wondering if you want to balance it out with some splitting / smooshing / etc.” (I have no idea what those would mean but you get the idea.)

I have no doubt that a thoughtful, extensive and shared vocabulary of this kind would elevate our profession. It would be a concrete vehicle for the transmission and development of our shared expertise in designing mathematical experiences.

This notion has some antecedents.[1] First, there are the passes at articulating what makes a problem pedagogically valuable. On the math blogosphere, see discussions by Avery Pickford, Breedeen Murray, and Michael Pershan. (Edit 1/21: I knew Dan had one of these too.) I also would like to believe that there is a well-developed discussion on this topic in academic print journals, although I am unaware of it. (A google search turned up this methodologically odd but interesting-seeming article about biomed students. Is it the tip of the iceberg? Is anyone reading this acquainted with the relevant literature?)

Also, I know a few other actual words that fit into the category “specialized vocabulary to discuss math tasks and problems.” I forget where I first ran into the word problematic in this context – possibly in the work of Cathy Twomey-Fosnot and Math in the City – but that’s a great word. It means that the problem feels authentic and vital; the opposite of contrived. I also forget where I first heard the word grabby (synonymous with Pershan’s hooky, and not far from how Dan uses perplexing) to describe a math problem – maybe from the lips of Justin Lanier? But, once you know it it’s pretty indispensible. Jo Boaler, by way of Dan Meyer, has given us the equally indispensable pseudocontext. So, the ball is already rolling.

When Cody shared his ideas, Yvonne and I speculated that the folks responsible for the PCMI problem setsBowen Kerins and Darryl Yong, and their friends at the EDC – have some sort of internal shared vocabulary of problem design, since they are masters. They were giving a talk today, so I went, and asked this question. It wasn’t really the setting to get into it, but superficially it sounded like yes. For starters, the PCMI’s problem sets (if you are not familiar with them, click through the link above – you will not be sorry) all contain problems labeled important, neat and tough. “Important” means accessible, and also at the center of connections to many other problems. Darryl talked about the importance of making sure the “important” problems have a “low threshold, high ceiling” (a phrase I know I’ve heard before – anyone know where that comes from?). He said that Bowen talks about “arcs,” roughly meaning, mathematical themes that run through the problem sets, but I wanted to hear much more about that. Bowen, are you reading this? What else can you tell us?

Most of these words share with Cody’s coinages the quality of being catchy / natural-language-feeling. They are not jargony. In other words, they are inclusive rather than exclusive.[2] It is possible for me to imagine that they could become a shared vocabulary of our whole profession.

So now what I really want to ultimately happen is for a whole bunch of people (Cody, Yvonne, Bowen, you, me…) to put in some serious work and to write a book called A Critical Language for Mathematical Problem Design, that catalogues, organizes and elucidates a large and supple vocabulary to describe the design of mathematical problems and tasks. To get this out of the completely-idle-fantasy stage, can we do a little brainstorming in the comments? Let’s get a proof of concept going. What other concepts for thinking about task design can you describe and (jargonlessly) name?

I’m casting the net wide here. Cody’s “jamming” and “thwarting” are verbs describing ways that problems can interrupt the rote application of methods. “Problematic” and “grabby” are ways of describing desirable features of problems, while “pseudocontext” is a way to describe negative features. Bowen and Darryl’s “important/neat/tough” are ways to conceptualize a problem’s role in a whole problem set / course of instruction. I’m looking for any word that you could use, in any way, when discussing the design of math tasks. Got anything for me?

[1]In fairness, for all I know, somebody has written a book entitled A Critical Language for Mathematical Task Design. I doubt it, but just in case, feel free to get me a copy for my birthday.

[2]I am taking a perhaps-undeserved dig here at a number of in-many-ways-wonderful curriculum and instructional design initiatives that have a lot of rich and deep thought about pedagogy behind them but have really jargony names, such as Understanding by Design and Cognitively Guided Instruction. (To prove that an instructional design paradigm does not have to be jargony, consider Three-Acts.) I feel a bit ungenerous with this criticism, but I can’t completely shake the feeling that jargony names are a kind of exclusion: if you really wanted everybody to use your ideas, you would have given them a name you could imagine everybody saying.

Dispatches from the Learning Lab: Partial Understanding

So here’s another one that I suppose is kind of obvious, but nonetheless feels like big, important news to me:

It’s possible to only partly understand what somebody else is saying.

Let me be more specific. When you’re explaining something to me, it’s possible for me to get some idea from it in a clear way, to the point where my understanding registers on my face, but nonetheless the other 7 ideas you were describing I have no idea what you’re talking about.


I am a 9th grader in your Algebra I class. You’re teaching me about linear functions. You are explaining to the class how to find the y-intercept of a linear function, in slope-intercept form, given that the slope is 4 and the point (6,11) lies on the line. You explain that the equation has the form y=mx+b and that because we know the point (6,11) is on the line, that this point satisfies the equation. Thus you write

11=4\cdot 6+b

on the board. At this point I recognize that we are trying to find b and that we have an easy single-variable linear equation to solve. My face lights up and you take mental note of my engagement. Maybe you even ask for the y-intercept, and since I recognize that this must be b I calculate 11-24 = -13 and raise my hand.

Meanwhile, I have only the vaguest sense of the meaning of the phrase “y-intercept.” I have literally no understanding of why I should expect the equation to have the form y=mx+b. I have a nagging feeling of dissatisfaction ever since you substituted (6,11) into the equation because I thought x and y were supposed to be the variables but now it looks like b is the variable. Most importantly, I do not understand that the presence of the point on the line implies that its coordinates satisfy the equation of the line and conversely, because on a very basic level I don’t understand what the graph of the function is a picture of. This has been bothering me ever since we started the unit, when you had me plug in a bunch of x values into some equations and obtain corresponding y values, graph them, and then draw a solid line connecting the three or four points. Why am I drawing these lines? What are they pictures of?

Occasionally, I’ve asked a question aimed at getting clarity on some of these basic points. “How did you know to put the 6 and 11 into the equation?” But because I can’t be articulate about what I don’t understand, since I don’t understand it, and you can’t hear what I’m missing in my questions because the the theory is complete and whole in your mind, these attempts come to the same unsatisfying conclusion every time. You explain again; I frown; you explain a different way; I say, “I don’t understand.” You, I, and everyone else grow uncomfortable as the impasse continues. Eventually, you offer some thought that has something in it for me to latch onto, just as I latched onto solving for b before. Just to dispel the tension and let you get on with your job, I say, “Ah! Yes, I understand.”


This example is my attempt to translate a few experiences I’ve had this semester into the setting of high school. The behavior of the student in that last paragraph was typical of me in these situations, though it would be atypical from a high school student, drawing as it does on the resources of my adulthood and educator background to self-advocate, to tolerate awkwardness, even to be aware that my understanding was incomplete. Still, often enough I ended up copping out as the student does above, understanding one of the 8 things that were going on, and latching onto it just so I could allow myself, the teacher and the class to move on gracefully. Conversations with other students indicated that my sense of incomplete understanding was entirely typical, even if my self-advocacy was not.

The take-home lesson is two-fold. Point one is about the limitations of explaining as a method of teaching. Point two is about the limitations of trusting your students’ (verbal or implied) response to your (verbal or implied) question, Do you understand?

The basic answer (as you can tell from the example) is, No, I don’t.

Now I myself love explaining and have done a great deal of it as a teacher. I fancy myself an extremely clear and articulate explainer. But it couldn’t be more abundantly clear, from this side of the desk, how limited is the experience of being explained to. I mean, actually it’s a great, key, important way to learn, but only in small doses and when I’m ready for it, when the groundwork for what you have to say has been properly set.

I am somewhat chastened by this. I am thinking back self-consciously to times when I’ve explained my students’ ears off rather than, in the immortal words of Shawn Cornally, “lay off and let them fucking think for a second.” It’s like I was too taken with the clarity and beauty of the formulation I was offering, or in too much of a hurry to let them work through what they had to work through, or in all likelihood both, to see that more words weren’t going to do any good. Beyond this, I’m thinking back on the faith I’ve put in my ability to read students’ level of understanding from their faces. I maintain that I’m way better at this than my professors, but I don’t think I’ve had enough respect for how you can understand a small part of something and have that feel like a big enough deal to say, and mean, “Oh I get it.” Or to understand a tiny part of something and use that as cover for not understanding the rest.

Partial Illumination for the Chords-of-an-Ellipse Problem

Back in July, Sam Shah described a beautiful and haunting problem he had encountered at PCMI:

Put n evenly spaced points on a unit circle, with one point at (1,0). Then draw chords from this point to all the other points. Then multiply the chords’ lengths. What do you get?

Mimi Yang drew some great illustrations of the problem, and Mr. Ho made a very slick geogebra applet that illustrates and gets data at the same time.

Sam mentioned an extension problem:

Scale the circle vertically by a factor of \sqrt{5}. Scale all the chords too. What is the product of the lengths now?

Mr. Ho outdid himself by creating a geogebra applet for this one too. In fact, he lets you scale the ellipse in all kinds of ways.

It’s really this extension that’s the subject of this post. But first…

Background on the original problem

If you haven’t bumped into this problem yet and it’s not obvious to you how to solve it, you might want to experiment a bit, if nothing else then with Mr. Ho’s applets. I am going to talk about the problem’s results, which are awesome, so I’m warning you now that if the problem is new to you but you read on, I’m going to steal some of your fun.

People posted solutions in the comments to Sam’s original problem statement. The solutions fell into 2 categories:

1) Direct calculation of the product using trigonometry. These methods were able to produce a surprising conjecture about the product of lengths for n points, but were not able to prove it.

2) Recognition that if the points are treated as numbers in the complex plane, they are precisely the nth roots of unity, followed by exploitation of the algebra of the nth roots of unity. These methods were able to prove the conjecture. (For example: Andrew’s comment at Sam’s original post; gasstationwithoutpumps has his/her (?) own post on the subject.)

This is as far as I can get before stating the result; so – SPOILER ALERT.

For n points on the circle, the product of all these chords, most of which have irrational lengths, is a positive integer. Not any positive integer, but n itself.

Here is the roots-of-unity proof, which I’ve done my best to render in a way that’s accessible even if you’ve never worked with roots of unity before. You do need to be familiar with the geometry of complex numbers though.

If you interpret the n equally spaced points on the circle as complex numbers, then they are precisely the nth roots of 1: \omega, \omega^2, \dots, \omega^{n-1} where \omega is the first of the numbers you find if you go around the circle counterclockwise after you leave the positive real axis. (Think about what happens when you multiply this number by itself n times, to convince yourself of this.) Since the point from which all the chords emanate is the number 1, the product of the lengths is the absolute value of the product of the complex numbers (1-\omega), (1-\omega^2), \dots, (1-\omega^{n-1}). This product has a startlingly elegant form, which can be seen by noticing that this product is exactly the product (z-\omega)(z-\omega^2) \dots (z-\omega^{n-1}) evaluated at z=1; and that this product is exactly the monic polynomial that has \omega, \omega^2, \dots, \omega^{n-1} as roots. What are they roots of? Oh right, unity. In other words, they are all zeros of the polynomial z^n-1. This polynomial also has 1 itself as a root (because 1 is an nth root of 1 too), so you have to divide it by a factor of z-1 in order to get the polynomial that has only the omegas as roots.



And now, evaluation of this at z=1 gives you 1+1+…+1 = n. The product of the chord lengths is the absolute value of this, but this is a positive real number so its absolute value is itself. Thus the product of the chord lengths is n. QED.

The ellipse extension

So everybody who produced a proof to the original problem did it by using the algebra of the nth roots of unity. The first thing that grabbed me about the ellipse extension is that while it is obviously closely related to the original problem, it immediately destroys the linchpin of this method. As soon as the circle gets stretched up to the ellipse, the points are no longer nth roots of unity!

The next thing that grabbed me about it was Tom’s comment on Sam’s original post:

I created my own Geogebra applet to investigate the problem with the ellipse (as you have it written I think). The products turn out to be very interesting:

The lengths of the cords are as follows for certain n:

n length
2 2=2*1
3 6=3*2
4 12=4*3
5 25= 5*5
6 48=6*8
7 91=7*13
8 168= 8*21
9 306=9*34
10 550=10*55

So the conjecture would then be

n n*F_n

where F_n is the n’th Fibonacci number.


(As an aside, the type of reaction I had to Tom’s comment is something we should be cultivating in students. Observation of an unexpected pattern leads naturally to a feverish search for an explanation. This is why it’s so important not to treat a pattern noticed as an established fact: this kills the students’ natural wonderment about the why of the pattern.)

Mr. Ho confirmed Tom’s calculations with his own applet. He had a look at other scale factors besides \sqrt{5} but didn’t find anything. At this point, the problem officially had itself ensconced semi-permanently in the back of my brain. How (the hell) are the frickin Fibonacci numbers arising? If you know something about the Fibonacci numbers, you know that they’re related to the golden ratio, which is related to \sqrt{5}, so that kind of makes sense, but c’mon now, only kind of. It’s not like the connection is jumping out at me.

Still, I couldn’t justify spending a lot of time on the problem. I try to stick with math problems that fit into a program of study I’ve given myself, and this one didn’t obviously do that. Nonetheless, it was lodged in my head firmly enough for me to describe it to my excellent colleague Japheth Wood (News from the Math Wizard). The next day, he sent me the following in an email:

Update on the stretched diagonals problem. I wrote a short program in
Python to do some calculations, and I found out the following
patterns, when sqrt{5} is replaced by sqrt{4a+1}:

Conjecture: The product of the diagonals of the stretched n-gon is P_n
= n*T_n, where T_n is an integer sequence defined by:
T_1 = T_2 = 1 and T_{n+2} = T_{n+1} + a*T_n
* This is true in the case of the non-stretched circle: a = 0, and
this recurrence gives P_n = n * 1, which is known.
* This seems true in the PCMI case. 5 = 4*1 + 1, so a = 1, and the
recurrence is Fibonacci.
* The other cases where a is an integer seem true by data I’ve collected.
* The conjecture even seems true when a is not an integer.


Okay, now the time had obviously come for me to give this problem some serious love. Japheth had just a) exploded onto a completely new level my sense that there’s a lot going on here, and b) given me enough of a direction that I knew some real elbow grease would pay off. I spent quite a few hours this weekend with the problem. What I found is still nowhere near a proof, but it strengthens Japheth’s conjecture and suggests some lines for further investigation. Here’s what I did:

First I found some closed forms for Japheth’s recursive sequence T_n:

Closed form #1: Using a standard method involving encoding the recursion T_{n+2}=T_{n+1}+aT_n into a matrix and then diagonalizing the matrix (which I learned – like all the linear algebra I know – from Michael Artin’s Algebra, chapters 3 and 4), I found that


where u and v are the two roots of the polynomial x^2-x-a. I didn’t end up using this result but I think it’s very pretty. I’m sure it’s well-known but it’s new to me.

Closed form #2: I got a different formula by directly calculating the first few values of T_n in terms of a, and looking for patterns:


I put this in a table for ease of calculation and pattern-searching:

n T_n= 1 a a^2 a^3 a^4
1 1
2 1
3 1 1
4 1 2
5 1 3 1
6 1 4 3
7 1 5 6 1
8 1 6 10 4
9 1 7 15 10 1

This is just a downward-slanted Pascal’s triangle! (Look at what the recursion T_{n+2}=T_{n+1}+aT_n does to a pair of rows to get the next row; this tells you why.) It follows that

T_n=\displaystyle\sum_{k=0}^{\left \lfloor (n-1)/2 \right \rfloor} \binom{n-k-1}{k} a^k

The next thing I did was to reason as follows: in the original circle problem, all the power of the roots-of-unity method came from knowing the polynomial x^n-1 that has these numbers as roots. Now that the circle has been stretched to an ellipse, the points no longer represent roots of unity; but if I could find the polynomial that had these numbers as roots…

I assumed that a would be such that \sqrt{4a+1} would be a real number, so that using it as the vertical stretch factor would make geometric sense. I put no other restrictions on a. Then I just started calculating:

n=2: The points are at 1 and -1; no change when you stretch. The polynomial is g(z)=(z-1)(z+1) = z^2-1.

n=3: The points are at 1 and \frac{-1 \pm \sqrt{3} \mathrm{i}}{2}. Vertical stretching by a factor of \sqrt{4a+1} makes the latter two into \frac{-1 \pm \sqrt{4a+1}\sqrt{3} \mathrm{i}}{2}. So, multiply:

g(z)=(z-1) \left ( z- \left ( \frac{-1 + \sqrt{4a+1}\sqrt{3} \mathrm{i}}{2} \right ) \right ) \left ( z- \left ( \frac{-1 - \sqrt{4a+1}\sqrt{3} \mathrm{i}}{2} \right ) \right )

When all is said and done, this simplifies to

g(z) = z^3 + 3az - (3a+1)

And so on. I made it to n=8. Let me tell you, n=7 was a dirty job. I looked for ways to make things easier – for example, I wrote the points as \cos(2\pi k/n)+\mathrm{i}\sqrt{4a+1}\sin(2\pi k/n) so that I could take advantage of some symmetry, cross-cancellation and trig identities, and for n=7 I found the polynomial that has \cos(2\pi k/7) as roots, which then gave me some symmetric expressions in the \cos(2\pi k/7)s that showed up in the final product, without needing to actually calculate the value of each \cos(2\pi k/7). But still. That case alone had to have taken me at least 2 hours. I messed up several times, thankfully in ways that contradicted each other. But I got my data:

n g(z)
1 z-1
2 z^2-1
3 z^3+3az-(3a+1)
4 z^4+4az^2-(4a+1)
5 z^5+5az^3+5a^2z-(5a^2+5a+1)
6 z^6+6az^4+9a^2z^2-(9a^2+6a+1)
7 z^7+7az^5+14a^2z^3+7a^3z-(7a^3+14a^2+7a+1)
8 z^8+8az^6+20a^2z^4+16a^3z^2-(16a^3+20a^2+8a+1)

Clear patterns are emerging in the structure of these polynomials, what with how
a) the powers of z are falling by 2 and of a are rising by 1
b) the constant term is the negative of the sum of the coefficients of the other terms

Together, these mean we could really regard the whole thing as a sum of terms of the form


The last thing to do is figure out what’s going on with the coefficients. The lead coefficient is always 1, the next (provided it contains a z) is counting numbers, and the next (again, provided z is involved) – the sequence 5, 9, 14, 20 – is increasing by counting numbers. This is an echo of the Pascal’s Triangle, so I looked for an expression for these numbers in terms of chooses. I found one: the k+1th coefficient of the nth polynomial (i.e. the coefficient of the a^kz^{n-2k} term) seems to be

\dbinom{n-k+1}{k} - \dbinom{n-k-1}{k-2}

So, without further ado –


* If you take the unit circle and place n evenly-spaced points along it, the first at (1,0)
* And you then scale everything vertically by a factor of \sqrt{4a+1}, where a is any real number such that \sqrt{4a+1} is real
* And you then regard each of these scaled points as a complex number,
* The monic polynomial g(z) with these numbers as roots will be equal to

\displaystyle\sum_{k=0}^{\left \lfloor (n-1)/2 \right \rfloor} \left [ \dbinom{n-k+1}{k} - \dbinom{n-k-1}{k-2} \right ] a^k (z^{n-2k} - 1)

If this conjecture is correct, Japheth’s conjecture follows. (For those of you who like that sort of thing, how does it follow?) So proving this conjecture would settle all the outstanding questions (e.g. it would prove the Fibonacci pattern in the \sqrt{4a+1}=\sqrt{5} case).

But I have no idea how to prove it. Why are the powers of a showing up? How about the chooses? What’s going on here?


(Darryl Young and Bowen Kerins, the writers of the PCMI problem set, are invited to give the rest of us hints!)