In short
An inverse trigonometric function takes a ratio as input and returns an angle as output. Because trig functions are periodic and hit each value many times, you must restrict their domains to make them one-to-one before inverting. The restricted version is called the principal value branch, and it gives a unique angle for each input. The six inverse trig functions — \sin^{-1}, \cos^{-1}, \tan^{-1}, \cot^{-1}, \sec^{-1}, \csc^{-1} — each have their own domain, range, and graph.
You know that \sin 30° = \frac{1}{2}. Now turn the question around: which angle has a sine of \frac{1}{2}?
If you are working with trigonometric equations, the answer is "infinitely many angles" — 30°, 150°, 390°, -210°, and so on. Every angle of the form n \cdot 180° + (-1)^n \cdot 30° works. That is the general solution.
But sometimes you do not want infinitely many answers. Sometimes you want the answer — a single, definite angle. When you type \sin^{-1}(0.5) into a calculator, it does not return a list of infinitely many numbers. It returns one number: 30° (or \frac{\pi}{6} in radian mode). The calculator has made a choice.
That choice — which one angle to return — is the entire subject of this article. The function that makes this choice is called the inverse sine function, written \sin^{-1} or \arcsin. It takes a number (a ratio) as input and returns an angle. But to make that function well-defined, you have to be very careful about which angle it is allowed to return. Get that wrong, and you get contradictions. Get it right, and you have a precise tool that converts ratios back into angles.
Why you cannot just "invert" sine directly
Here is the problem in its sharpest form.
A function has an inverse only if it is one-to-one: each output comes from exactly one input. Consider f(x) = x^2. It is not one-to-one because f(3) = 9 and f(-3) = 9 — two different inputs give the same output. So "f^{-1}(9)" is ambiguous: is it 3 or -3?
The sine function has the same problem, but worse. Not just two inputs give the same output — infinitely many do. The function \sin x takes the value \frac{1}{2} at x = \frac{\pi}{6}, \frac{5\pi}{6}, \frac{13\pi}{6}, -\frac{7\pi}{6}, \ldots
If you try to define \sin^{-1}\left(\frac{1}{2}\right) as "the angle whose sine is \frac{1}{2}," the definition is ambiguous — there are infinitely many such angles. A function cannot return infinitely many values. A function must return exactly one.
The solution: restrict the domain of sine to an interval where it is one-to-one, and then invert only that restricted version.
Choosing the right restriction: the principal value branch
Which interval should you pick? You need a piece of the sine curve where:
- Sine is one-to-one (no two inputs give the same output — the curve does not double back on itself).
- The piece covers the entire range [-1, 1] (so that \sin^{-1}(y) is defined for every y between -1 and +1).
For sine, the natural choice is \left[-\frac{\pi}{2}, \frac{\pi}{2}\right]. On this interval, the sine curve goes from \sin(-\frac{\pi}{2}) = -1 up to \sin(\frac{\pi}{2}) = 1, hitting every value between -1 and 1 exactly once. It is strictly increasing, never doubling back.
This restricted piece is called the principal value branch of sine. When you invert it, you get the inverse sine function.
The three main inverse trig functions
Inverse sine ($\sin^{-1}$ or $\arcsin$)
Domain: [-1, 1]. Range: \left[-\frac{\pi}{2}, \frac{\pi}{2}\right].
Inverse cosine ($\cos^{-1}$ or $\arccos$)
Domain: [-1, 1]. Range: [0, \pi].
Inverse tangent ($\tan^{-1}$ or $\arctan$)
Domain: (-\infty, \infty). Range: \left(-\frac{\pi}{2}, \frac{\pi}{2}\right).
Notice the patterns. For each function, the chosen range is an interval where the original trig function is one-to-one and covers its full range of values:
- Sine increases from -1 to 1 on \left[-\frac{\pi}{2}, \frac{\pi}{2}\right], so that is the range of \sin^{-1}.
- Cosine decreases from 1 to -1 on [0, \pi], so that is the range of \cos^{-1}.
- Tangent increases from -\infty to +\infty on \left(-\frac{\pi}{2}, \frac{\pi}{2}\right), so that is the range of \tan^{-1}.
Why [0, \pi] for cosine instead of [-\frac{\pi}{2}, \frac{\pi}{2}]? Because cosine is not one-to-one on [-\frac{\pi}{2}, \frac{\pi}{2}] — it takes the value \frac{\sqrt{3}}{2} at both \frac{\pi}{6} and -\frac{\pi}{6}. On [0, \pi], cosine is strictly decreasing, so each value appears exactly once.
Reading the notation
The notation \sin^{-1} x is standard in Indian textbooks (NCERT, RD Sharma, and most JEE materials). The notation \arcsin x means exactly the same thing and is more common in some international contexts. Both are used interchangeably.
One critical warning: \sin^{-1} x does not mean \frac{1}{\sin x}. The superscript -1 here denotes the inverse function, not a reciprocal. The reciprocal of \sin x is \csc x (or (\sin x)^{-1} if you must use a power notation). This is an unfortunate ambiguity in mathematical notation, and the only way to handle it is to memorise the convention.
The graphs
The graph of an inverse function is obtained by reflecting the graph of the original function across the line y = x. What was horizontal becomes vertical; what was the x-axis becomes the y-axis.
Graph of \sin^{-1} x
Take the principal branch of sine (from -\frac{\pi}{2} to \frac{\pi}{2}) and reflect it across y = x. The result is an S-shaped curve that starts at (-1, -\frac{\pi}{2}) and ends at (1, \frac{\pi}{2}).
Key features:
- Domain: [-1, 1]. You cannot compute \sin^{-1}(2) — it does not exist.
- Range: \left[-\frac{\pi}{2}, \frac{\pi}{2}\right].
- Odd function: \sin^{-1}(-x) = -\sin^{-1}(x).
- Increasing throughout its domain.
Graph of \cos^{-1} x
Key features:
- Domain: [-1, 1].
- Range: [0, \pi].
- Decreasing throughout its domain. Larger inputs give smaller angles.
- Not odd, not even. But there is a symmetry: \cos^{-1}(-x) = \pi - \cos^{-1}(x).
Graph of \tan^{-1} x
Key features:
- Domain: all real numbers (-\infty, \infty).
- Range: \left(-\frac{\pi}{2}, \frac{\pi}{2}\right) — open interval, since the values \pm\frac{\pi}{2} are never actually reached.
- Odd function: \tan^{-1}(-x) = -\tan^{-1}(x).
- Increasing throughout its domain.
- Horizontal asymptotes at y = \frac{\pi}{2} and y = -\frac{\pi}{2}.
Basic properties
These identities follow directly from the definitions and are used constantly.
Cancellation identities
For \sin^{-1}:
- \sin(\sin^{-1} x) = x for all x \in [-1, 1].
- \sin^{-1}(\sin x) = x only when x \in \left[-\frac{\pi}{2}, \frac{\pi}{2}\right].
The first identity says: take a number, find the angle whose sine is that number, then take the sine of that angle — you get back where you started. This always works within the domain.
The second identity is the one that trips people up. If x = \frac{5\pi}{6}, then \sin \frac{5\pi}{6} = \frac{1}{2}, and \sin^{-1}\left(\frac{1}{2}\right) = \frac{\pi}{6}, not \frac{5\pi}{6}. You do not get back to where you started because \frac{5\pi}{6} is outside the range \left[-\frac{\pi}{2}, \frac{\pi}{2}\right].
The general rule: \sin^{-1}(\sin x) returns the angle in \left[-\frac{\pi}{2}, \frac{\pi}{2}\right] that has the same sine as x. If x is already in that interval, the answer is x. If not, you have to find the equivalent angle in the principal range.
The same logic applies to cosine and tangent:
- \cos^{-1}(\cos x) = x only when x \in [0, \pi].
- \tan^{-1}(\tan x) = x only when x \in \left(-\frac{\pi}{2}, \frac{\pi}{2}\right).
The complementary identity
Here is why. Let \sin^{-1} x = \alpha. Then \sin \alpha = x and \alpha \in \left[-\frac{\pi}{2}, \frac{\pi}{2}\right].
Now \cos\left(\frac{\pi}{2} - \alpha\right) = \sin \alpha = x. The angle \frac{\pi}{2} - \alpha lies in [0, \pi] (check: when \alpha = -\frac{\pi}{2}, you get \pi; when \alpha = \frac{\pi}{2}, you get 0). So \frac{\pi}{2} - \alpha is the angle in [0, \pi] whose cosine is x — which is exactly the definition of \cos^{-1} x.
Therefore \cos^{-1} x = \frac{\pi}{2} - \alpha = \frac{\pi}{2} - \sin^{-1} x, and rearranging gives \sin^{-1} x + \cos^{-1} x = \frac{\pi}{2}.
This identity says that the "inverse sine angle" and the "inverse cosine angle" for the same input always add up to a right angle. Geometrically, this is just the fact that the two acute angles in a right triangle sum to \frac{\pi}{2}.
Negative argument identities
For \sin^{-1}: Let \sin^{-1} x = \alpha. Then \sin \alpha = x, so \sin(-\alpha) = -x, and since -\alpha \in [-\frac{\pi}{2}, \frac{\pi}{2}], we get \sin^{-1}(-x) = -\alpha = -\sin^{-1} x. This is exactly the statement that \sin^{-1} is an odd function.
For \cos^{-1}: Let \cos^{-1} x = \alpha, so \cos \alpha = x and \alpha \in [0, \pi]. Then \cos(\pi - \alpha) = -\cos \alpha = -x, and \pi - \alpha \in [0, \pi]. So \cos^{-1}(-x) = \pi - \alpha = \pi - \cos^{-1} x.
For \tan^{-1}: the argument is the same as for \sin^{-1}, using the fact that tangent is an odd function.
Computing specific values
Here is a table of the most commonly needed values. Every entry follows from the standard trig table and the principal value ranges.
| x | \sin^{-1} x | \cos^{-1} x |
|---|---|---|
| 0 | 0 | \frac{\pi}{2} |
| \frac{1}{2} | \frac{\pi}{6} | \frac{\pi}{3} |
| \frac{1}{\sqrt{2}} | \frac{\pi}{4} | \frac{\pi}{4} |
| \frac{\sqrt{3}}{2} | \frac{\pi}{3} | \frac{\pi}{6} |
| 1 | \frac{\pi}{2} | 0 |
| -\frac{1}{2} | -\frac{\pi}{6} | \frac{2\pi}{3} |
| -1 | -\frac{\pi}{2} | \pi |
Notice how each row confirms \sin^{-1} x + \cos^{-1} x = \frac{\pi}{2}.
| x | \tan^{-1} x |
|---|---|
| 0 | 0 |
| \frac{1}{\sqrt{3}} | \frac{\pi}{6} |
| 1 | \frac{\pi}{4} |
| \sqrt{3} | \frac{\pi}{3} |
| -1 | -\frac{\pi}{4} |
Example 1: Find $\sin^{-1}\left(\sin \frac{7\pi}{6}\right)$
This is the classic "trap" problem. The angle \frac{7\pi}{6} is not in the range of \sin^{-1}, so the answer is not \frac{7\pi}{6}.
Step 1. Compute \sin \frac{7\pi}{6}.
Why: \frac{7\pi}{6} is in the third quadrant, where sine is negative. The reference angle is \frac{7\pi}{6} - \pi = \frac{\pi}{6}.
Step 2. Now compute \sin^{-1}\left(-\frac{1}{2}\right).
You need the angle in \left[-\frac{\pi}{2}, \frac{\pi}{2}\right] whose sine is -\frac{1}{2}. That angle is -\frac{\pi}{6}.
Why: \sin\left(-\frac{\pi}{6}\right) = -\frac{1}{2}, and -\frac{\pi}{6} is in the principal range \left[-\frac{\pi}{2}, \frac{\pi}{2}\right].
Step 3. Verify using the negative argument identity.
\sin^{-1}\left(-\frac{1}{2}\right) = -\sin^{-1}\left(\frac{1}{2}\right) = -\frac{\pi}{6}.
Why: \sin^{-1} is odd, so \sin^{-1}(-x) = -\sin^{-1}(x). This confirms the answer from Step 2.
Result: \sin^{-1}\left(\sin \frac{7\pi}{6}\right) = -\frac{\pi}{6}.
The unit circle makes the answer visible. The angle \frac{7\pi}{6} and the angle -\frac{\pi}{6} land on the same horizontal line (same y-coordinate). The inverse sine function always picks the one on the right half of the circle — the one in \left[-\frac{\pi}{2}, \frac{\pi}{2}\right].
Example 2: Simplify $\tan^{-1} 1 + \cos^{-1}\left(-\frac{1}{2}\right) + \sin^{-1}\left(-\frac{1}{2}\right)$
Step 1. Evaluate each term separately.
\tan^{-1} 1: The angle in \left(-\frac{\pi}{2}, \frac{\pi}{2}\right) whose tangent is 1 is \frac{\pi}{4}.
Why: \tan \frac{\pi}{4} = 1, and \frac{\pi}{4} is in the principal range of \tan^{-1}.
Step 2. Evaluate \cos^{-1}\left(-\frac{1}{2}\right).
\cos^{-1}\left(-\frac{1}{2}\right) = \pi - \cos^{-1}\left(\frac{1}{2}\right) = \pi - \frac{\pi}{3} = \frac{2\pi}{3}.
Why: the identity \cos^{-1}(-x) = \pi - \cos^{-1}(x) converts a negative argument into a positive one. And \cos^{-1}\frac{1}{2} = \frac{\pi}{3} from the standard table.
Step 3. Evaluate \sin^{-1}\left(-\frac{1}{2}\right).
\sin^{-1}\left(-\frac{1}{2}\right) = -\sin^{-1}\left(\frac{1}{2}\right) = -\frac{\pi}{6}.
Why: \sin^{-1} is odd. And \sin^{-1}\frac{1}{2} = \frac{\pi}{6}.
Step 4. Add.
Common denominator is 12:
Why: pure arithmetic with fractions. The key was knowing each inverse trig value first, then combining.
Result: \tan^{-1} 1 + \cos^{-1}\left(-\frac{1}{2}\right) + \sin^{-1}\left(-\frac{1}{2}\right) = \frac{3\pi}{4}.
This kind of problem tests whether you know the principal value ranges cold. There is no trick beyond knowing that \cos^{-1}(-\frac{1}{2}) = \frac{2\pi}{3} (not -\frac{\pi}{3} or \frac{4\pi}{3}) and \sin^{-1}(-\frac{1}{2}) = -\frac{\pi}{6} (not \frac{11\pi}{6}).
Common confusions
-
"\sin^{-1} x = \frac{1}{\sin x}." No. \sin^{-1} x is the inverse function, not the reciprocal. The reciprocal of \sin x is \csc x. The notations look dangerously similar, which is why many textbooks prefer \arcsin over \sin^{-1}.
-
"\sin^{-1}(\sin x) = x always." Only when x is already in \left[-\frac{\pi}{2}, \frac{\pi}{2}\right]. If x = \pi, then \sin \pi = 0, and \sin^{-1}(0) = 0 \neq \pi.
-
"\cos^{-1}\left(-\frac{1}{2}\right) = -\frac{\pi}{3}." The range of \cos^{-1} is [0, \pi], so it never returns a negative angle. The correct answer is \frac{2\pi}{3}.
-
"The graph of \sin^{-1} goes from 0 to \pi." That is \cos^{-1}. The graph of \sin^{-1} goes from -\frac{\pi}{2} to \frac{\pi}{2}. Confusing the ranges of the two functions is the single most common error on exams involving inverse trig.
-
"\tan^{-1} x has the same domain as \sin^{-1} x." No. \sin^{-1} x accepts only inputs in [-1, 1], but \tan^{-1} x accepts any real number. You can compute \tan^{-1}(1000) — it is approximately \frac{\pi}{2}, just a hair less.
Going deeper
If you came here to understand what inverse trig functions are, how to read their graphs, and how to evaluate them at standard values, you have it. The rest of this section covers the remaining three inverse trig functions and the connection to trigonometric equations.
The other three: \cot^{-1}, \sec^{-1}, \csc^{-1}
The three reciprocal trig functions also have inverse functions, though they appear less frequently.
\cot^{-1} x: Domain: (-\infty, \infty). Range: (0, \pi). The principal branch of cotangent is taken on (0, \pi), where it is strictly decreasing from +\infty to -\infty.
\sec^{-1} x: Domain: (-\infty, -1] \cup [1, \infty). Range: [0, \pi] \setminus \left\{\frac{\pi}{2}\right\} — i.e., \left[0, \frac{\pi}{2}\right) \cup \left(\frac{\pi}{2}, \pi\right]. The gap at \frac{\pi}{2} exists because \sec \frac{\pi}{2} is undefined.
\csc^{-1} x: Domain: (-\infty, -1] \cup [1, \infty). Range: \left[-\frac{\pi}{2}, \frac{\pi}{2}\right] \setminus \{0\}. Similar reasoning: \csc 0 is undefined.
These three can always be converted to the main three:
So in practice, you can work entirely with \sin^{-1}, \cos^{-1}, and \tan^{-1} and convert when needed.
Connection to trigonometric equations
Inverse trig functions are the formal machinery behind the "principal solution" concept from Trigonometric Equations. When you solve \sin x = \frac{1}{2} and write the general solution as x = n\pi + (-1)^n \frac{\pi}{6}, the value \frac{\pi}{6} is precisely \sin^{-1}\frac{1}{2} — the principal value.
More generally, the general solution of \sin x = a (for |a| \leq 1) can be written as:
The inverse function gives you the starting angle; the general solution formula generates the full infinite family from it. Without the inverse function, you would have to look up the principal value from a table or a calculator every time. With it, the answer is a self-contained algebraic expression.
Madhava and the infinite series for \arctan
Long before the notation \tan^{-1} existed, the Kerala mathematician Madhava of Sangamagrama (14th century) discovered the infinite series
valid for |x| \leq 1. Setting x = 1 gives
This is one of the earliest known formulas for computing \pi as an infinite series, predating European discovery of the same result by roughly 250 years. The series converges slowly — you need thousands of terms for even a few decimal places of \pi — but its mathematical beauty is undeniable. It says that \pi is hiding inside the inverse tangent function, which is hiding inside the geometry of the unit circle.
Where this leads next
You now know what inverse trig functions are, why their domains and ranges are what they are, and how to evaluate them. The natural next steps:
- Inverse Trigonometric Identities — sum formulas like \tan^{-1} a + \tan^{-1} b, composition identities, and simplification techniques.
- Trigonometric Equations — using inverse trig functions to express principal solutions in general solution formulas.
- Trigonometric Functions and Graphs — the forward direction, if you want to revisit the original trig functions before studying their inverses.
- Inverse Functions — the general theory of function inverses, of which inverse trig functions are a specific case.
- Graph Transformations — Reflections — why reflecting across y = x turns a function's graph into its inverse's graph.