You are here
Hometrigonometry
Primary tabs
trigonometry
1 Geometrical definitions
Trigonometry arose in ancient times out of attempts to measure the lengths of various lines associated with circles. For example, consider this diagram showing the upper right quadrant with a unit circle:
The point $B$, external to the circle with center $O$, determines a unique secant $BO$ and two tangent lines of equal length, one of which is $BA$. It thus also uniquely determines an angle $\alpha=\angle AOB$. The trigonometric functions $\sec$ and $\tan$ simply measure the ratio of the lengths of the secant and tangent lines to the radius of the circle.
Thus, in the diagram, $\tan\alpha=AB=AB:OA$. But triangles $OAB$ and $ODC$ are obviously similar, so $\tan\alpha$ is also equal to $CD:OD$, and we recover the usual currentday definition of $\tan$. Similarly, from the diagram, $\sec\alpha=OB=OB:OA$ and again, using similarity, we also have $\sec\alpha=OC:OD$, which is again the usual modern definition.
The cosecant, $\csc$, is the ratio of the secant to the other leg of the right triangle. Using the standard definitions today for $\sec$ and $\csc$, we see that this is correct:
$\csc(x)=\frac{\sec(x)}{\tan(x)}=\frac{1}{\cos(x)}\cdot\frac{cos(x)}{\sin(x)}=% \frac{1}{\sin(x)}$ 
The sine arose in attempts to measure the length of a chord traversing a given angle on a circle. Consider the following diagram:
Here, the arc $CE$ is the circular angle $\alpha$, which is also $\angle CFE$, $\angle COD$, and $\angle DOE$ (recall that a circumferential angle is the same measure as the arc subtended, while a central angle is twice the arc subtended). The length of the chord $CE$ is in fact $2\sin\alpha$. Presumably because central angles are easier to work with, the common definition was that of the halfchord, and one can see by considering $\triangle COD$ that in fact $CD$ is $\sin\alpha$ using the modern definitions.
The term “sine” has an interesting history. The Hindus gave the name jiva to the halfchord $CD$; the Arabs used (or created) the word jiba for this concept. When Robert of Chester, an early translator of alKhowarizmi’s Algebra, translated this word, he mistook it for a similar Arabic word, jaib, which means “bay” or “inlet”. As a result, he used the Latin term “sinus”, which also means bay or inlet.
To understand how the trigonometric functions are defined today, consider the right triangle $ABC$ below.
Noting that $0^{\circ}<\alpha<90^{\circ}$, we define the trigonometric functions sine, cosine, tangent, secant, cosecant and cotangent for angle $\alpha$ respectively as:
$\sin(\alpha)=\frac{BC}{CA},\qquad\qquad\csc(\alpha)=\frac{CA}{BC}$ 
$\cos(\alpha)=\frac{AB}{CA},\qquad\qquad\sec(\alpha)=\frac{CA}{AB}$ 
$\tan(\alpha)=\frac{BC}{AB},\qquad\qquad\cot(\alpha)=\frac{AB}{BC}$ 
We will discuss later how to extend these definitions to a broader set of values for $\alpha$.
Several identities follow directly from the definitions:

$\tan(\alpha)=\sin(\alpha)/\cos(\alpha)$.

$\sec(\alpha)=1/\cos(\alpha),\qquad\csc(\alpha)=1/\sin(\alpha),\qquad\cot(% \alpha)=1/\tan(\alpha)$.

$\sin(\alpha)=\cos(90^{\circ}\alpha),\qquad\tan(\alpha)=\cot(90^{\circ}\alpha% ),\qquad\sec(\alpha)=\csc(90^{\circ}\alpha)$.
The last property follows from the fact that $\angle ACB=90^{\circ}\alpha$.
The Pythagorean theorem states that $CA^{2}=AB^{2}+BC^{2}$ and thus
$1=\frac{CA^{2}}{CA^{2}}=\frac{AB^{2}+BC^{2}}{CA^{2}}=\left(\frac{AB}{CA}\right% )^{2}+\left(\frac{BC}{CA}\right)^{2}=(\cos(\alpha))^{2}+(\sin(\alpha))^{2}.$ 
It is customary to write $(\sin(\alpha))^{n}$, $(\cos(\alpha))^{n}$, etc. as $\sin^{n}(\alpha)$, $\cos^{n}(\alpha)$, etc. respectively, so the previous identity is usually written as
$\sin^{2}(\alpha)+\cos^{2}(\alpha)=1$ 
and it is known as the Pythagorean identity. Notice that the first three identities let us to express any expression involving trigonometric functions using only sines and cosines, whereas the Pythagorean identity lets us reduce it further, using only sines. This technique is sometimes used when proving trigonometric identities.
Two other identities that can be obtained from the Pythagorean theorem or from the Pythagorean identity are
$\tan^{2}(\alpha)+1=\sec^{2}(\alpha),\qquad 1+\cot^{2}(\alpha)=\csc^{2}(\alpha).$ 
2 Extending the domain
There are several approaches for extending the domain of the trigonometric functions so they are not restricted to angles between $0$ and $90^{\circ}$. We could use the angle sum identities below in order to calculate the value of the trigonometric functions outside the given range (for instance, $\sin(120^{\circ})$ could be found as $\sin(60^{\circ}+60^{\circ})$. Some other more formal approaches use power series expansion to define the functions for any real value (or even complex!), but we will use a more geometrical approach, using the unit circle.
Consider a unit circle centered at origin of the plane (that is, a circle of radius $1$ and center $(0,0)$), and draw a ray from the center making an angle of $\alpha$ with the horizontal axis, measured counterclockwise from the positive $x$axis. (This ray is the terminal ray of the angle.) Let $P$ be the point of intersection of this line with the circle, and $T$ the intersection with the line $x=1$. Finally, let us denote by $(x,y)$ the coordinates of $P$ and $(1,t)$ the coordinates of $T$.
Since the circle has radius $1$, the distance from origin to $P$ is also $1$ and thus
$\cos\alpha=\frac{x}{1}=x,\qquad\sin\alpha=\frac{y}{1}=y,$ 
in other words, the coordinates of $P$ are precisely $(\cos\alpha,\sin\alpha)$.
Now, the two right triangles made with the projections of $P$ and $T$ to the $x$axis are similar, so we have
$\tan\alpha=\frac{y}{x}=\frac{t}{1}=t,$ 
that is, the point $t$ has ordinate equal to $\tan\alpha$.
This formulation makes it much easier to generalize the definitions to arbitrary angles: $\cos\alpha,\sin\alpha$ are the coordinates of the intersection of the circle with the line through the origin making angle $\alpha$ with the positive $x$axis:
You may be wondering now why we put $\tan\alpha$ instead of $\tan\alpha$ if the analogy was to be continued. The reason is that we only need $\sin\alpha$ and $\cos\alpha$ to define all the other trigonometric functions, and given that
$\tan\alpha=\frac{\sin\alpha}{\cos\alpha}$ 
we use that identity to extend the domain of $\tan\alpha$. In the particular case of the drawing above, both $\cos\alpha$ and $\sin\alpha$ are negative, so their quotient must be positive, and that’s why we added a minus sign to $\tan\alpha$.
The geometrical approach lets us easily verify the following relations:
$\displaystyle\sin(180^{\circ}\alpha)=\sin\alpha,$  $\displaystyle\sin(180^{\circ}+\alpha)=\sin\alpha,$  $\displaystyle\sin(\alpha)=\sin\alpha$  
$\displaystyle\cos(180^{\circ}\alpha)=\cos\alpha,$  $\displaystyle\cos(180^{\circ}+\alpha)=\cos\alpha,$  $\displaystyle\cos(\alpha)=\cos\alpha$  
$\displaystyle\tan(180^{\circ}\alpha)=\tan\alpha,$  $\displaystyle\tan(180^{\circ}+\alpha)=\tan\alpha,$  $\displaystyle\tan(\alpha)=\tan\alpha$ 
3 Graphs
The graph of $\sin$ appears below. Note that the $x$axis is scaled not in degrees, but rather in radians.
The graph of $\cos$ is below. Note the similarity between the two graphs: the graphs of $\cos$ is just like the graph of $\sin$, but shifted by $90^{\circ}$ ($\pi/2$ radians). This precisely reflects the identity $\sin\alpha=\cos(\alpha90^{\circ})$.
Also notice that you can read off directly from the graphs the facts that
$\sin\alpha=\sin(\alpha),\qquad\cos\alpha=\cos(\alpha)$ 
4 Trigonometrical identities
There are literally thousands of trigonometric identities. Some of the most common (and most useful) are:
Sum and difference of angles
$\displaystyle\sin(x+y)$  $\displaystyle=\sin x\cos y+\cos x\sin y,$  $\displaystyle\ \sin(xy)$  $\displaystyle=\sin x\cos y\cos x\sin y,$  
$\displaystyle\cos(x+y)$  $\displaystyle=\cos x\cos y\sin x\sin y,$  $\displaystyle\ \cos(xy)$  $\displaystyle=\cos x\cos y+\sin x\sin y,$  
$\displaystyle\tan(x+y)$  $\displaystyle=\frac{\tan x+\tan y}{1\tan x\tan y},$  $\displaystyle\ \tan(xy)$  $\displaystyle=\frac{\tan x\tan y}{1+\tan x\tan y}.$ 
(There is a proof of angle sum identities, as well as a geometric derivation of addition formulas for sine and cosine, on this site).
Half and double angles
The double angle formulas are derived directly from the sum of angles formulas above. The halfangle formulas can then be derived from the double angle formulas by substituting $x/2$ for $x$ and simplifying, using the identity $sin^{2}+cos^{2}=1$.
$\displaystyle\sin(x/2)=\pm\sqrt{\frac{1\cos x}{2}},$  $\displaystyle\sin(2x)$  $\displaystyle=2\sin x\cos x,$  
$\displaystyle\cos(x/2)=\pm\sqrt{\frac{1+\cos x}{2}},$  $\displaystyle\cos(2x)$  $\displaystyle=\cos^{2}x\sin^{2}x=2\cos^{2}x1=12\sin^{2}x.$ 
Mathematics Subject Classification
4300 no label found5100 no label found4200 no label found33B10 no label found Forums
 Planetary Bugs
 HS/Secondary
 University/Tertiary
 Graduate/Advanced
 Industry/Practice
 Research Topics
 LaTeX help
 Math Comptetitions
 Math History
 Math Humor
 PlanetMath Comments
 PlanetMath System Updates and News
 PlanetMath help
 PlanetMath.ORG
 Strategic Communications Development
 The Math Pub
 Testing messages (ignore)
 Other useful stuff
Recent Activity
new question: Lorenz system by David Bankom
Oct 19
new correction: examples and OEIS sequences by fizzie
Oct 13
new correction: Define Galois correspondence by porton
Oct 7
new correction: Closure properties on languages: DCFL not closed under reversal by babou
new correction: DCFLs are not closed under reversal by petey
Oct 2
new correction: Many corrections by Smarandache
Sep 28
new question: how to contest an entry? by zorba
new question: simple question by parag
Sep 26
new question: Latent variable by adam_reith
Attached Articles
table of values for $\cos(x)$ by drini
table of values for $\tan(x)$ by drini
$\operatorname{arc tan}$ with two arguments by matte
exact trigonometry tables by stevecheng
goniometric formulas by Wkbj79
proof of angle sum identities by rspuzio
phase parametrization in $\cos(x+p)=\cos x$ by perucho
Comments
basic trigonometric entry
to my surprise there wasn't an entry defining and dealing and defining the elementary trigonometric functions, so I went and created this.
I know this entry is the kind of entry that will generate a long thread about comments, suggestions criticism and all that jazz, so why don't someone better sets up an asteroidmeta page for this so we can avoid polluting the page (and making it a lot larger and slower to load ;)
It's incomplete as it stands right now, but I'm making it world editable so anyone can work on it directly. Ah, and I'd like it to keep it as real trig mostly entru, so please do not dwell too much into complex analysis (perhaps just some mention and references)
f
G > H G
p \ /_  ~ f(G)
\ / f ker f
G/ker f
Re: basic trigonometric entry
Actually, THERE IS already an entry defining elementary trigonometric functions:
http://planetmath.org/encyclopedia/DefinitionsInTrigonometry.html
Re: basic trigonometric entry
Another entry defining geometrically the sine and cosine of all real numbers is
http://planetmath.org/encyclopedia/CyclometricFunctions.html.
Re: basic trigonometric entry
does cyclometric functions one really defines them?
it seems a bit circular to me if you want them to define sin, cos, etc
(unless you use the power series at the bottom to define arcsin first and then define sin as the inverse of arcsin, but then it gets all too contrived)
f
G > H G
p \ /_  ~ f(G)
\ / f ker f
G/ker f
Re: basic trigonometric entry
The first section of your entry is a repeat of "definitions in trigonometry" and as a matter of fact the previous entry is a little clearer in the exposition (for my taste). And that entry defines (i.e. they are in the "also defines" list) sine, cosine and the kind.
I like the idea of extending the topic and adding a bunch of things that "defs in trig. " is missing though, that is why I suggested making your entry a child of that one.
Alvaro
Re: basic trigonometric entry
Ok I'll attach it, but keep in mind I said the entry as it stands is far from complete, right now I'm redoing the picture so I can fill the correction you still have opened, and many other things need to be added
f
G > H G
p \ /_  ~ f(G)
\ / f ker f
G/ker f
Re: basic trigonometric entry
Thanks!
And again, thanks for adding this entry to the collection, I have also noticed that Planetmath was a little weak in trigonometry.
Alvaro
Re: basic trigonometric entry
As I see it, there really should be both definitions available. The definition in terms of sides and angles is uitable for beginners studying geometry and the definition in terms of power series and invers functions is more suitable for more advanced people.
Hoewever, this discussion should never have happened here, so I am going to build the wiki so we can carry out the discussion in a more suitable place.
Trigonometry discussion page
Please post all further discussion of this entry to the following webpage:
http://oddwiki.taoriver.net/wiki.pl/AsteroidMeta/Discussion_of_Trigonometry
Re: basic trigonometric entry
Maybe, it seems a bit circular. In fact the 'defining' of sine and cosine (for all real numbers) is only a subplot in explaining the old names arcsin and arccos  you will see it if you read anew the whole article. But possibly I shall improve the order of things there.  The power series here is not any definition of arcsin, but only a formula.
Jussi
Re: basic trigonometric entry
all I pointed is that I can't see the cyclometric entry defining the trig functions, just making use of them
f
G > H G
p \ /_  ~ f(G)
\ / f ker f
G/ker f
Re: basic trigonometric entry
All right =o)