If I have been given that $$\sin{x} = \frac{1 + \sqrt{5}}{4}$$ is there any way using which I can find out what the value of $x$ is? Sure I can remember the values, but I was curious whether if I knew the values of $\sin{\frac{\pi}{2}}, \sin{\frac{\pi}{4}}, \sin{\frac{\pi}{3}}$ I could somehow figure out what $x$ would be? Is there any intuition behind it?
-
6Related problem: https://math.stackexchange.com/questions/2811476/cool-way-of-finding-cos-left-frac-pi5-right – Gary Apr 22 '21 at 07:16
-
1If you know the value of the $\sin x$, you know the side lengths of some right-angled triangle with angle $x$. I would start there. – Arthur Apr 22 '21 at 07:20
-
1You can cheat first: https://www.wolframalpha.com/input/?i=arcsin++%28%281%2Bsqrt%285%29%29%2F4%29. Then try to derive it yourself. – Martin R Apr 22 '21 at 07:22
-
2Deriving it knowing the angle is doable, is there any way to reverse the process ? finding the angle from the given values, knowing some specific values ? – marks_404 Apr 22 '21 at 07:27
-
4There is no immediate connection with between this angle (which happens to be $3\pi/10$) and the other "special angles", because the denominator $10$ isn't a common denominator of linear combinations of those other angles. ... Now, what one comes to expect from expressions with $\sqrt{5}$ is that associated trig values probably involve $\pi/5$ (due to connections between the Golden Ratio and the regular pentagon). Thus, one might well expect $5x$ to have "nice" trig values; so you could expand, say, $\sin 5x$ in terms of $\sin x$ and see what happens. – Blue Apr 22 '21 at 07:29
3 Answers
Here is a tricky way to get at the solution without "knowing" it.
Start with $\sin x =(\sqrt5+1)/4$. Convert this to a quadratic equation by identifying the conjugate $(-\sqrt5+1)/4$ as the other root. Then since the sum of these two roots is $1/2$ and the product is $-1/4$, we identify the quadratic equation
$4\sin^2x-2\sin x-1=0$
Now multiply by $\cos x$ and apply the double angle formula $\sin 2x=2\sin x\cos x$:
$2\sin x\sin2x-\sin 2x-\cos x=0$
And now apply the sum-product relation $2\sin u\sin v =\cos (u-v)-\cos(u+v)$ (be careful with signs!), and note what happens to the $\cos x$ term: $\require{cancel}$
$\cancel{\cos x}-\cos 3x-\sin 2x\cancel{-\cos x}=0$
$\color{blue}{\cos 3x=-\sin 2x}$
So, from the complementary angle relations we must have one of the following:
$\color{blue}{3x+2x=(4n+3)\pi/2,n\in\mathbb{Z}}$
$\color{brown}{3x-2x=(4m+1)\pi/2,m\in\mathbb{Z}}$
If $x$ is to be a positive quantity strictly between $0$ and $\pi/2$ (radians), then only the first of these two relations gives a solution and it matches the required range only for $n=0$! Thereby $\color{blue}{x=3\pi/10}$ is selected among arguments between $0$ and $\pi/2$.
It is instructive to plot the solutions to the two complementary-angle relations.

The blue-colored rays correspond to the blue equation above, while the brown-colored ray is matched to the brown equation.
Four of the blue roots correspond to the original quadratic equation for $\sin x$. The second unique value of $\sin x$ from these roots is the second root of the quadratic equation, $(-\sqrt5+1)/4$, and it corresponds to $x=-\pi/10$. Thus we have retrieved $\sin^{-1}[(-1+\sqrt5)/4]=\pi/10$ along with $\sin^{-1}[(1+\sqrt5)/4]=3\pi/10$. Such a dual result would be expected from any method based on the quadratic equation.
We see that introducing the $\cos x$ factor completes a fivefold symmetric set of the blue roots. This offers a geometric/symmetry-related explanation why the $\cos x$ multiplier worked as well as it did. This multiplier also introduced a root not part of the symmetric group, which corresponds to the brown equation.
- 48,208
-
1You need to add a
\require{cancel}somewhere before the first use of\cancel. – user3733558 Apr 23 '21 at 21:30 -
2So that's what happened. As you can see I basically gave up. Thank you. Uh, is my math better than my syntax? :-S – Oscar Lanzi Apr 23 '21 at 21:42
-
1It took me ages to figure out the cancel thing. I still don't know why this is required, but that's how it works on Math.SE, oh well. As for the maths, nice and informative argument, upvoted :) – user3733558 Apr 23 '21 at 21:59
-
How did you come up with the idea of multiplying the equation by cos x? Btw, the conjugate has a wrong sign – Jonas De Schouwer Apr 23 '21 at 22:18
-
I wanted to get rid of the non-unit coefficients, if I don't the complementary-angle relations don't work. Multiplying by $\cos x$ and putting $\sin 2x=2\sin x\cos x$ is a step in that direction. Sign corrected, typo. – Oscar Lanzi Apr 23 '21 at 22:26
Let $\cos y = \frac{\sqrt5+1}4$. Then $$\cos 2y =2\cos^2 y-1= 2\left(\frac{\sqrt5+1}4\right)^2-1=\frac{\sqrt5-1}4$$ $$\cos 4y =2\cos^2 2y-1= -\frac{\sqrt5+1}4= -\cos y= \cos(\pi\pm y)$$ The $+$ sign leads to $y=\frac\pi3$ while the $-$ sign leads to $y=\frac\pi5$. But then the value above for $\cos 2y$ is positive so we must have $y<\frac\pi4$. Therefore, $y=\frac\pi5$ is selected and $x=\frac\pi2-\frac\pi5=\frac{3\pi}{10}$.
- 48,208
- 120,125
-
2Another elegant solution from trig identities most students know. +1. – Oscar Lanzi Apr 23 '21 at 23:59
-
Note the correction I added. $\cos 4y=-\cos y$ leads to another root which is "obviously" wrong but properly must be proven so. – Oscar Lanzi Apr 28 '21 at 22:27
Hint for this particular problem. The fraction $$ \frac{1 + \sqrt{5}}{4} $$ is half the golden ratio $$ \varphi = \frac{1 + \sqrt{5}}{2} $$ which strongly suggests that a regular pentagon is involved. It's no surprise that the answer $3\pi/10$ has denominator divisible by $5$.
You can work out by inscribing this pentagon picture from cut-the-knot in a unit circle:
in which
$$
\frac{DE}{EX} =
\frac{EX}{XY} =
\frac{UV}{XY} =
\frac{EY}{EX} =
\frac{BE}{AE} = \varphi
$$
- 103,433