Sign of trigonometric functions

Ask whether sin⁡200∘\sin 200^\circ is positive or negative and the unit circle answers instantly, without a calculator: it depends only on which of the four quadrants the angle's terminal side lands in. Every one of the six trigonometric functions is built from the coordinates (cos⁡x,sin⁡x)(\cos x, \sin x) of a point moving around a circle of radius 11, and coordinates flip sign exactly as they do in ordinary coordinate geometry — positive-positive in Quadrant I, negative-positive in Quadrant II, and so on. That single observation, dressed up in the mnemonic "All Silver Tea Cups," lets a student read off the sign of sin⁡x\sin x, cos⁡x\cos x, tan⁡x\tan x, cosec⁡x\operatorname{cosec} x, sec⁡x\sec x and cot⁡x\cot x in any quadrant without memorising six separate rules, and it is the missing piece that turns the identity sin⁡2x+cos⁡2x=1\sin^2x+\cos^2x=1 from "two possible values" into one definite answer.

The core insight

The sign of each trigonometric function in a given quadrant is exactly the sign of the coordinate ratio that defines it, since cos⁡x\cos x and sin⁡x\sin x are the xx- and yy-coordinates of a unit-circle point and every other function is built from their ratios or reciprocals.

The core idea

On the unit circle, a point P(a,b)P(a,b) at angle xx has cos⁡x=a\cos x = a and sin⁡x=b\sin x = b — the coordinates of PP are literally the values of cosine and sine. As xx sweeps through a full revolution, PP visits all four quadrants in turn, and aa and bb change sign exactly as xx- and yy-coordinates do anywhere in coordinate geometry.

  • Quadrant I (0<x<π20 < x < \tfrac{\pi}{2}): a>0, b>0a>0,\ b>0, so sin⁡x>0\sin x>0 and cos⁡x>0\cos x>0 — and since every other function is a ratio or reciprocal of these two, all six functions are positive here.
  • Quadrant II (π2<x<π\tfrac{\pi}{2} < x < \pi): a<0, b>0a<0,\ b>0, so sin⁡x>0\sin x>0 and cosec⁡x>0\operatorname{cosec} x>0, while cos⁡x,sec⁡x,tan⁡x,cot⁡x\cos x, \sec x, \tan x, \cot x are negative.
  • Quadrant III (π<x<3π2\pi < x < \tfrac{3\pi}{2}): a<0, b<0a<0,\ b<0, so sin⁡x\sin x and cos⁡x\cos x are both negative; but tan⁡x=sin⁡xcos⁡x\tan x = \dfrac{\sin x}{\cos x} is a negative divided by a negative, hence positive, and cot⁡x\cot x is positive too. sec⁡x\sec x and cosec⁡x\operatorname{cosec} x stay negative.
  • Quadrant IV (3π2<x<2π\tfrac{3\pi}{2} < x < 2\pi): a>0, b<0a>0,\ b<0, so cos⁡x>0\cos x>0 and sec⁡x>0\sec x>0, while sin⁡x,cosec⁡x,tan⁡x,cot⁡x\sin x, \operatorname{cosec} x, \tan x, \cot x are negative.

This pattern is captured by the mnemonic "All Silver Tea Cups" (equivalently "All Sin Tan Cos"), read quadrant by quadrant from Quadrant I onward:

QI: All positive,QII: Sin, cosec positive,QIII: Tan, cot positive,QIV: Cos, sec positive.\text{QI: All positive},\quad \text{QII: Sin, cosec positive},\quad \text{QIII: Tan, cot positive},\quad \text{QIV: Cos, sec positive}.

where: each entry lists the two functions that are positive in that quadrant; every other function there is negative. At the quadrantal angles 0,π2,π,3π2,2π,…0, \tfrac{\pi}{2}, \pi, \tfrac{3\pi}{2}, 2\pi,\dots one of sin⁡x\sin x or cos⁡x\cos x equals 00, so any function that divides by it becomes undefined — these boundary angles belong to no open quadrant and carry no "sign" at all.

Figure
The unit circle divided into four quadrants each labelled with its two positive trigonometric functions following the All Silver Tea Cups mnemonic
All Silver Tea Cups: what is positive in each quadrant QI All positive QII Sin, cosec + QIII Tan, cot + QIV Cos, sec + P(a, b) x y
Derivation ASigns of the six functions from the unit circle

The goal is to derive, from nothing but the unit-circle definitions cos⁡x=a\cos x = a, sin⁡x=b\sin x = b and the quadrant signs of the coordinates (a,b)(a,b), the full sign of every one of the six trigonometric functions in Quadrant III, and to show why this reasoning generalises to give the "All Silver Tea Cups" pattern in every quadrant.

Step 1. Fix the quadrant and read off the coordinate signs. Take xx in Quadrant III, so π<x<3π2\pi < x < \tfrac{3\pi}{2}. The point P(a,b)=(cos⁡x,sin⁡x)P(a,b)=(\cos x,\sin x) lies in the third quadrant of the coordinate plane, where by definition both coordinates are negative: a<0a<0 and b<0b<0. Hence cos⁡x<0\cos x<0 and sin⁡x<0\sin x<0 directly, with no further work needed for these two.

Step 2. Get the reciprocal functions for free. sec⁡x=1cos⁡x\sec x = \dfrac{1}{\cos x} has the same sign as cos⁡x\cos x because dividing 11 (positive) by a negative number gives a negative number; so sec⁡x<0\sec x<0. Likewise cosec⁡x=1sin⁡x\operatorname{cosec} x = \dfrac{1}{\sin x} shares sine's sign, so cosec⁡x<0\operatorname{cosec} x<0. This is the general rule: a reciprocal never changes sign, so sec⁡x\sec x always matches cos⁡x\cos x and cosec⁡x\operatorname{cosec} x always matches sin⁡x\sin x.

Step 3. Combine the two negatives for tangent. tan⁡x=sin⁡xcos⁡x=ba=negativenegative\tan x = \dfrac{\sin x}{\cos x} = \dfrac{b}{a} = \dfrac{\text{negative}}{\text{negative}}. A negative divided by a negative is positive, so tan⁡x>0\tan x>0 in Quadrant III — the one place where a naive glance ("both parts are negative") tempts a student to guess wrongly.

Step 4. Finish with cotangent and generalise. cot⁡x=1tan⁡x\cot x = \dfrac{1}{\tan x} shares tangent's sign by the same reciprocal rule as Step 2, so cot⁡x>0\cot x>0. Repeating this four-step process — read the coordinate signs of the quadrant, assign sin⁡x,cos⁡x\sin x,\cos x, take reciprocals for cosec⁡x,sec⁡x\operatorname{cosec} x,\sec x, and divide for tan⁡x,cot⁡x\tan x,\cot x — reproduces every row of the ASTC table without memorising it directly; only the signs of aa and bb in each quadrant need to be recalled.

Worked example

Problem. Given tan⁡x=−512\tan x = -\dfrac{5}{12} with xx in Quadrant IV, find sin⁡x\sin x and cos⁡x\cos x.

Step 1. Use the identity 1+tan⁡2x=sec⁡2x1+\tan^2x=\sec^2x to bring in cos⁡x\cos x. Compute sec⁡2x=1+(−512)2=1+25144=169144\sec^2x = 1+\left(-\dfrac{5}{12}\right)^2 = 1+\dfrac{25}{144}=\dfrac{169}{144}, so sec⁡x=±1312\sec x=\pm\dfrac{13}{12}.

Step 2. Fix the sign using the quadrant. In Quadrant IV, cos⁡x>0\cos x>0 (from the ASTC pattern above), and sec⁡x\sec x shares cosine's sign, so sec⁡x=1312\sec x=\dfrac{13}{12} and hence cos⁡x=1213\cos x=\dfrac{12}{13}.

Step 3. Recover sin⁡x\sin x from tan⁡x=sin⁡x/cos⁡x\tan x=\sin x/\cos x, i.e. sin⁡x=tan⁡x⋅cos⁡x\sin x=\tan x\cdot\cos x:

sin⁡x=(−512)(1213)=−513.\sin x = \left(-\frac{5}{12}\right)\left(\frac{12}{13}\right) = -\frac{5}{13}.

Step 4. Check against the quadrant. In Quadrant IV, sin⁡x\sin x must be negative — and −513-\tfrac{5}{13} is negative, confirming the answer. sin⁡x=−513\sin x = -\dfrac{5}{13} and cos⁡x=1213\cos x = \dfrac{12}{13}.

Competitive stretch

If instead only sin⁡x<0\sin x<0 and cos⁡x<0\cos x<0 were given, the quadrant sign chart shows sin⁡x<0\sin x<0 occurs in QIII and QIV, while cos⁡x<0\cos x<0 occurs in QII and QIII — the only quadrant common to both lists is QIII, so tan⁡x=negativenegative>0\tan x = \dfrac{\text{negative}}{\text{negative}}>0 and cosec⁡x=1negative<0\operatorname{cosec} x = \dfrac{1}{\text{negative}}<0, exactly as Step 3 of the derivation predicts.

Common traps

  • Applying ASTC only to sine, cosine and tangent and forgetting the reciprocal functions: cosec⁡x\operatorname{cosec} x always matches sin⁡x\sin x's sign, sec⁡x\sec x always matches cos⁡x\cos x's sign, and cot⁡x\cot x always matches tan⁡x\tan x's sign.
  • Picking the wrong quadrant when only one condition is given. A single condition like sin⁡x<0\sin x<0 is true in two quadrants (III and IV); a second condition on cos⁡x\cos x or tan⁡x\tan x is needed to pin down exactly one.
  • Concluding "negative times negative is negative" for tangent in Quadrant III, when a ratio of two negatives is actually positive — this is the single most common sign slip in the whole topic.
  • Ignoring that quadrantal angles are undefined for some functions. At x=π2x=\tfrac{\pi}{2}, tan⁡x\tan x and sec⁡x\sec x are undefined; at x=πx=\pi, cot⁡x\cot x and cosec⁡x\operatorname{cosec} x are undefined — these boundary points have no quadrant and no sign.

Ask the tutor: still mixing up which functions are positive in Quadrant III versus Quadrant IV — want a quick timed ASTC drill from the AI tutor?

Why does All Silver Tea Cups work — is it just a coincidence of naming?

No, it is a direct summary of the unit-circle definitions. Sine and cosine are the coordinates of a point going around a circle, so their signs follow ordinary coordinate-geometry rules quadrant by quadrant. Every other function — tangent, cotangent, secant, cosecant — is either a ratio or a reciprocal of sine and cosine, so its sign is forced once sine and cosine's signs are known. "All Silver Tea Cups" is just a memory hook for the pattern that falls out of that logic; you can always rederive it from the coordinate signs if you forget the mnemonic.

If I am told sin x is positive, can I say for sure which quadrant x is in?

Not by itself. sin⁡x>0\sin x>0 is true in both Quadrant I and Quadrant II, so one condition alone never fixes a unique quadrant. You need a second piece of information — the sign of cos⁡x\cos x or tan⁡x\tan x, or an explicit restriction on the range of xx — to narrow it down to exactly one quadrant. This is exactly why sign questions on this topic almost always give two conditions together.

Does the sign of a trigonometric function depend on the value of sin squared x plus cos squared x equals 1?

Not directly — that identity only fixes the magnitude relationship between sin⁡x\sin x and cos⁡x\cos x, never their signs. For example, knowing cos⁡x=1213\cos x=\tfrac{12}{13} tells you sin⁡2x=1−144169=25169\sin^2x=1-\tfrac{144}{169}=\tfrac{25}{169}, so sin⁡x=±513\sin x=\pm\tfrac{5}{13} — but only the quadrant tells you which sign to keep. The identity supplies the number; the quadrant supplies the sign.

Before this

Where this leads