College Algebra introduced \mathbb C to finish what real numbers leave undone: every polynomial equation gets a solution once i=\sqrt{-1} is allowed. This course needs that same \mathbb C for a different reason — starting in chapter 06, "vector space" routinely means \mathbb C^n rather than \mathbb R^n, because that is the only setting in which "every operator has an eigenvalue" is unconditionally true. This section is a fast, focused refresher on exactly the complex-number facts that setting depends on: conjugation, modulus, and how they interact with vectors and inner products — not a full reintroduction to \mathbb C itself.
Think of a complex number as a point on a map with two coordinates: real east-west, imaginary north-south. Conjugating just reflects that point across the east-west axis. Flip the sign of the north-south coordinate, and leave everything else alone. The modulus is nothing but the ordinary distance from that point back to the origin, found with the Pythagorean theorem the exact same way it always has been. Multiplying a number by its own reflection cancels the flip and leaves a plain distance squared behind.
A complex number is a pair of real numbers written a+bi, added and multiplied like ordinary numbers subject to the one extra rule i^2=-1. Three facts carry the whole course. First, the conjugate \overline{z} = a-bi of z=a+bi satisfies z\overline{z} = a^2+b^2 = |z|^2, a real, nonnegative number — this single identity is what lets |z| (the modulus) serve as a genuine notion of length for a complex number, the same job |x| does for a real one. Second, conjugation distributes over every arithmetic operation: \overline{z_1+z_2}=\overline{z_1}+\overline{z_2} and \overline{z_1 z_2} = \overline{z_1}\,\overline{z_2}, which is exactly why a real matrix's complex eigenvalues always arrive in conjugate pairs (chapter 06) — applying a real matrix and then conjugating gives the same result as conjugating first. Third, extending a vector or matrix into \mathbb C^n means every entry is now a complex number, and the standard inner product on \mathbb C^n has to conjugate one side — \langle \mathbf u,\mathbf v\rangle = \sum u_i\overline{v_i} — precisely so that \langle \mathbf v,\mathbf v\rangle = \sum|v_i|^2 \ge 0 stays real and nonnegative, which is chapter 05's very first departure from the plain dot product.
Multiplication as rotate-and-scale
Drag either point — the product's angle is always the SUM of the two angles, and its distance from the origin is always the PRODUCT of the two distances.
Loading interactive…
A conjugate pair, added
Drag the point — watch its sum with its own reflection always land exactly on the real axis, with zero imaginary part.
Loading interactive…
Recipe Compute a complex inner product \langle u, v\rangle Use when: you need \langle u, v\rangle for vectors with complex-number entries, where the plain dot product no longer applies Goal: the single complex number \langle u,v\rangle = \sum u_k\overline{v_k}, with the conjugate on the correct side
Quick recipe
-
1
This course fixes one convention for every complex inner product: conjugate the second vector's entries, never the first, so \langle \mathbf u,\mathbf v\rangle=\sum u_k\overline{v_k}.
Example
For \mathbf u=(2+i,\ 1-i) and \mathbf v=(1+i,\ 2), it is \mathbf v whose entries get conjugated, not \mathbf u.
Why: swapping which vector is conjugated silently swaps \langle \mathbf u,\mathbf v\rangle into \langle \mathbf v,\mathbf u\rangle, its own conjugate, not the same number.
-
2
Flip the sign of the imaginary part of each entry of the vector named in Step 1, leaving the other vector's entries untouched.
Example
\overline{1+i}=1-i and \overline{2}=2, so conjugating \mathbf v=(1+i,\ 2) gives (1-i,\ 2).
Why: conjugating termwise is exactly what keeps \langle \mathbf v,\mathbf v\rangle a genuine nonnegative length once the two vectors coincide.
-
3
Multiply each entry of \mathbf u by the matching conjugated entry from Step 2, then sum every product.
Example
(2+i)(1-i)=2-2i+i-i^2=3-i, and (1-i)(2)=2-2i; adding gives 3-i+2-2i=5-3i.
- the two vectors are different — the sum can be any complex number, real or not.
- the two vectors are the same vector — the sum must come out real and nonnegative, checked in Step 4.
Why: this sum, not a plain dot product, is the definition fixed for every complex vector space this course studies.
-
4
Whenever the two vectors coincide, confirm the result has no leftover imaginary part and is not negative.
Example
For \mathbf v=(1+i,\ 2): \langle \mathbf v,\mathbf v\rangle=(1+i)(1-i)+(2)(2)=2+4=6, real and positive.
⚡ Shortcut
For a single complex number z (not a vector), this recipe collapses to the modulus computation: \langle z,z\rangle = z\overline z = |z|^2, so |z|=\sqrt{z\overline z}. For z=3-4i: \overline z=3+4i, z\overline z=9+16=25, so |z|=5.
Why: a negative or non-real result here would mean a conjugation slip earlier, since every genuine squared length must be real and nonnegative.
Common mistakes
- Conjugating the first vector's entries instead of the second, reversing the answer into its own conjugate.
- Forgetting to conjugate at all, which can produce a negative "squared length" for \langle \mathbf v,\mathbf v\rangle.
- A sign slip multiplying two complex numbers, especially forgetting i^2=-1.
- Skipping the real-and-nonnegative check on \langle \mathbf v,\mathbf v\rangle, the fastest way to catch a conjugation mistake.
Worked example
Computing a modulus and a conjugate
-
1flip only the sign of the imaginary part — the real part is untouched
This is the recipe's shortcut case above: a single complex number instead of a vector, computing \overline z and using it to find |z|. For z = 3-4i, find \overline z and |z|. \overline z = 3 - (-4)i = 3+4i
flip only the sign of the imaginary part — the real part is untouched
-
2|z| = \sqrt{z\overline z} = \sqrt{(3-4i)(3+4i)} = \sqrt{9-16i^2} = \sqrt{9+16}
-
3\boxed{\overline z = 3+4i, \qquad |z| = \sqrt{25} = 5}
Loading interactive…
Worked example
A complex inner product, not the plain dot product
-
1conjugate the SECOND vector's entries, not the first — this convention is fixed for the whole course
This runs the full recipe above, Steps 1 through 3, on a new pair of vectors. Compute \langle \mathbf u, \mathbf v\rangle in \mathbb C^2 for \mathbf u = (1+i,\ 2) and \mathbf v = (1,\ i), using \langle \mathbf u,\mathbf v\rangle = \sum u_k\overline{v_k}. \langle \mathbf u, \mathbf v\rangle = (1+i)\overline{(1)} + (2)\overline{(i)}
conjugate the SECOND vector's entries, not the first — this convention is fixed for the whole course
-
2
\overline{1} = 1 and \overline{i} = -i, so: \langle \mathbf u,\mathbf v\rangle = (1+i)(1) + (2)(-i) = 1+i-2i
-
3\boxed{\langle \mathbf u, \mathbf v\rangle = 1-i}
Reversing the order gives \langle \mathbf v,\mathbf u\rangle = \overline{1-i} = 1+i — conjugate symmetry, not plain symmetry, which is exactly the departure chapter 05 builds the rest of its theory on.
Loading interactive…
Worked example
Why \\langle v, v\\rangle must conjugate to stay real
-
1
Confirm \langle \mathbf v,\mathbf v\rangle is real and nonnegative for \mathbf v = (1, 2i), and show what goes wrong without the conjugate. \text{With conjugation: } \langle \mathbf v,\mathbf v\rangle = (1)\overline{(1)} + (2i)\overline{(2i)} = (1)(1) + (2i)(-2i)
-
2
(2i)(-2i) = -4i^2 = 4, so \langle \mathbf v,\mathbf v\rangle = 1+4 = 5 — real and positive, as required.
-
3\boxed{\text{Without conjugating: } (1)(1)+(2i)(2i) = 1+4i^2 = 1-4 = -3}
A negative "squared length" is nonsense for measuring size — exactly why the complex inner product conjugates one argument and the real dot product never needed to.
Loading interactive…
Check your understanding (counts toward your progress)
A real matrix's complex eigenvalues always arrive in conjugate pairs because
Right on try — credit. Right — the answer had already been shown, so no credit this time. Next try is worth Answer shown after three tries — this one no longer earns credit. You chose to see the answer — this one earns no credit.
This section's whole point is the opposite — a real matrix like a rotation can have genuinely non-real eigenvalues.
A single complex number is a single number, not a pair — the PAIRING is a consequence of real coefficients, not a definitional fact.
The characteristic polynomial's degree equals the matrix size, not always 2 — this pairing fact holds at every degree.
Why
If A\mathbf v=\lambda\mathbf v with A real, conjugating both sides gives A\overline{\mathbf v}=\overline\lambda\,\overline{\mathbf v} (A is unaffected since it's real) — so \overline\lambda is automatically an eigenvalue too.
Conjugate both sides of A\mathbf v = \lambda\mathbf v and use that A itself has only real entries.
For z=-3+4i, find \overline z + z (this sum is always real — enter its value).
Right on try — credit. Right — the answer had already been shown, so no credit this time. Next try is worth Answer shown after three tries — this one no longer earns credit. You chose to see the answer — this one earns no credit.
Why
\overline z = -3-4i, so \overline z+z = -6+0i = -6.
Correct answer
Also accepted:
Why
The imaginary parts of z and its conjugate cancel; only twice the real part survives.
Find |z| for z = 6-8i.
Right on try — credit. Right — the answer had already been shown, so no credit this time. Next try is worth Answer shown after three tries — this one no longer earns credit. You chose to see the answer — this one earns no credit.
Why
|z| = \sqrt{6^2+(-8)^2} = \sqrt{36+64} = \sqrt{100} = 10.
Correct answer
Also accepted:
Why
Use |z| = \sqrt{a^2+b^2}.
For z = a+bi, the conjugate \overline z is formed by
Right on try — credit. Right — the answer had already been shown, so no credit this time. Next try is worth Answer shown after three tries — this one no longer earns credit. You chose to see the answer — this one earns no credit.
That negates z entirely (-z), not the conjugate — the real part of a conjugate never changes.
The real part stays exactly the same under conjugation; only the imaginary part's sign flips.
Swapping the parts is a different operation entirely and isn't what conjugation means.
Why
\overline{a+bi} = a-bi — only the imaginary part's sign changes.
Practice: Complex arithmetic practice (4 problems · doesn’t affect your progress)
Compute the modulus |z| when z = 5+12i.
Right on try — credit. Right — the answer had already been shown, so no credit this time. Next try is worth Answer shown after three tries — this one no longer earns credit. You chose to see the answer — this one earns no credit.
Why
|z|=\sqrt{5^2+12^2}=\sqrt{25+144}=\sqrt{169}=13.
Correct answer
Also accepted:
Why
Find \overline z for z = -2+7i.
Right on try — credit. Right — the answer had already been shown, so no credit this time. Next try is worth Answer shown after three tries — this one no longer earns credit. You chose to see the answer — this one earns no credit.
Any equivalent form counts — 2x+3, 3 + 2·x, … Use the keypad below the field for fractions, roots, and other symbols.
Why
Only the imaginary part's sign flips: \overline{-2+7i} = -2-7i.
Correct answer
Also accepted:
Why
Compute \langle \mathbf u,\mathbf v\rangle's real part for \mathbf u=(1,\,2i), \mathbf v=(1,\,i) in \mathbb C^2.
Right on try — credit. Right — the answer had already been shown, so no credit this time. Next try is worth Answer shown after three tries — this one no longer earns credit. You chose to see the answer — this one earns no credit.
Why
(1)(1)+(2i)(\overline i) = 1+(2i)(-i) = 1+2 = 3, with imaginary part 0.
Correct answer
Also accepted:
Why
\langle \mathbf u,\mathbf v\rangle = u_1\overline{v_1}+u_2\overline{v_2} = (1)(1)+(2i)(-i).
Verify z\overline z is real by computing it for z=1-3i.
Right on try — credit. Right — the answer had already been shown, so no credit this time. Next try is worth Answer shown after three tries — this one no longer earns credit. You chose to see the answer — this one earns no credit.
Any equivalent form counts — 2x+3, 3 + 2·x, … Use the keypad below the field for fractions, roots, and other symbols.
Why
(1-3i)(1+3i) = 1-9i^2 = 1+9 = 10.
Correct answer
Also accepted:
Why
Where this shows up: conjugate pairs, everywhere
Every real matrix's complex eigenvalues arrive in conjugate pairs (chapter 06), every Hermitian form conjugates one of its two arguments (chapter 04), and every complex inner product conjugates one vector (chapter 05) — three appearances of the exact same algebraic fact reviewed here: conjugation is what keeps a real, symmetric-flavored theory alive once complex numbers are unavoidable.
Where this goes next
Signal processing and quantum mechanics
A discrete Fourier transform's output is a list of complex numbers whose modulus reports a signal's strength at each frequency and whose conjugate symmetry (for a real-valued input signal) is exactly what lets half the spectrum be discarded as redundant. Quantum mechanics goes further still: a state vector's entries are complex by necessity, and the probability of measuring any particular outcome is literally |\langle \psi_k, \psi\rangle|^2 — a modulus-squared of a complex inner product, not a metaphor for one.