# The Story about The Determinant

Hi folks,

I always had some difficulties understanding the concept of determinant. Whenever I read about it, and that was usually from Linear Algebra books, it was always described with a lot of "mathematical formality" and abstractness, and although I could follow and understand the definition and properties, the concept generally seemed to me like coming out of the blue.

"The best intuition" I have about determinant is that it measures how much does space get scaled by certain transformation. This intuition helped me a lot to understand some other concepts involving determinant, for example deriving the proof and explaining to myself why non-trivial solutions of Homogeneous System of Linear Equations occur when determinant of a matrix of the corresponding transformation equals 0. But, although I can prove this intuition holds for 2-dimension case, I find no way to generalize it for $n$ dimensions.

I feel that concept of determinant is really important to understand because it shows up in many branches of mathematics. I would really appreciate if someone could tell me a story about the determinant. I say story because it doesn't need to be highly mathematical and should aim to answer the following two questions:

• Why did mathematicians introduced the concept of determinant? Why they needed it at the first place?

• Based on that, how they derived the formula and generalized it for any number of dimensions?

Note by Uros Stojkovic
2 years, 9 months ago

This discussion board is a place to discuss our Daily Challenges and the math and science related to those challenges. Explanations are more than just a solution — they should explain the steps and thinking strategies that you used to obtain the solution. Comments should further the discussion of math and science.

When posting on Brilliant:

• Use the emojis to react to an explanation, whether you're congratulating a job well done , or just really confused .
• Ask specific questions about the challenge or the steps in somebody's explanation. Well-posed questions can add a lot to the discussion, but posting "I don't understand!" doesn't help anyone.
• Try to contribute something new to the discussion, whether it is an extension, generalization or other idea related to the challenge.

MarkdownAppears as
*italics* or _italics_ italics
**bold** or __bold__ bold
- bulleted- list
• bulleted
• list
1. numbered2. list
1. numbered
2. list
Note: you must add a full line of space before and after lists for them to show up correctly
paragraph 1paragraph 2

paragraph 1

paragraph 2

[example link](https://brilliant.org)example link
> This is a quote
This is a quote
    # I indented these lines
# 4 spaces, and now they show
# up as a code block.

print "hello world"
# I indented these lines
# 4 spaces, and now they show
# up as a code block.

print "hello world"
MathAppears as
Remember to wrap math in $$ ... $$ or $ ... $ to ensure proper formatting.
2 \times 3 $2 \times 3$
2^{34} $2^{34}$
a_{i-1} $a_{i-1}$
\frac{2}{3} $\frac{2}{3}$
\sqrt{2} $\sqrt{2}$
\sum_{i=1}^3 $\sum_{i=1}^3$
\sin \theta $\sin \theta$
\boxed{123} $\boxed{123}$

Sort by:

I don't have a story to tell, but I do have an anecdote. Suppose you have a matrix and a vector, such that the matrix pre-multiplied by the vector yields a vector of zeroes:

$\begin{pmatrix} a&b\\ c&d \end{pmatrix} \begin{pmatrix} x \\ y \end{pmatrix} = \begin{pmatrix} 0 \\ 0 \end{pmatrix}$

Suppose none of the elements $(a,b,c,d,x,y)$ are zero, for simplicity. What must the relationships between $a,b,c,d$ be?

$a x + b y = 0 \\ c x + d y = 0$

Rearranging gives:

$y = -\frac{a}{b} x$

Plugging into the second equation:

$c x - d \frac{a}{b} x = 0 \\ c - d \frac{a}{b} = 0 \\ b c - d a = 0 = a d - b c$

The quantity $(a d - b c)$ is well-known as the two-by-two matrix determinant. The determinant must be zero for the pre-multiplication of a vector by a matrix to yield a vector of zeros. It's almost as though the determinant is the matrix equivalent of a scalar magnitude.

- 2 years, 9 months ago

Proof which I derived for any $n \times n$ matrix and which I presented in my work.

"I will briefly prove here that non-trivial solutions of any homogeneous system of linear equations written in vector form $A\vec{x} = 0$ has non-trivial solutions when $\det(A) = 0.$ Let $i^{th}$ column entries of $A$ form vector $u_{i}$ so that: $A = \begin{bmatrix} a_{11} & a_{12} & \cdots & a_{1n} \\ a_{21} & a_{22} & \cdots & a_{2n} \\ \vdots & \vdots & \ddots & \vdots \\ a_{n1} & a_{n2} & \cdots & a_{nn} \end{bmatrix}= \begin{bmatrix} | & | & & | \\ u_{1} & u_{2} & \cdots & u_{n} \\ | & | & & | \\ \end{bmatrix}$

Then, by definition of matrix multiplication, we have:

$A\vec{x} = x_{1}u_{1} + x_{2}u_{2} + \cdots + x_{n}u_{n} = \theta,$

where $x_{i}$ is a scalar and, by conditions, at least one of them is non-zero. The above expression indicates that vectors $u_{1}, u_{2}, \cdots, u_{n}$ are linearly dependent. Ultimately, the determinant of a matrix formed by linearly dependent vectors is known to be zero. This comes from the fact that linear dependence implies that we can write some vector $u_{i}$ as a linear combination of other $n-1$ vectors ie. that $u_{i} \in \mathbb{R}^{k}$ where $k\leq n-1$. Hence, the transformation defined by that matrix would squish the space to a lower dimension which is captured by zero determinant."

This comes from the intuition I mentioned: determinant is the quantitative measure of space transformation defined by a certain matrix.

You wrote about $\mathbb{R}^{2}$ case when calculations are pretty straightforward, but what about $\mathbb{R}^{3}$, $\mathbb{R}^{4}$, $\mathbb{R}^{n}$ space? I guess that the most elegant way would be to derive a proof using induction. But I'm wondering if that's actually possible?

- 2 years, 9 months ago

Here's my idea of generalization of the formula by induction. Assume we have $n$ linear homogeneous equations and $n$ variables: \begin{aligned}a_{11}x_{1} + a_{12}x_{2} + \cdots + a_{1n}x_{n} &= \sum_{k = 1}^{n}a_{1k}x_{k} = 0, \\ a_{21}x_{1} + a_{22}x_{2} + \cdots + a_{2n}x_{n} &= \sum_{k = 1}^{n}a_{2k}x_{k} = 0, \\ \vdots ~~~~~~~~~~~~~~~~&\vdots \\ a_{n1}x_{1} + a_{n2}x_{2} + \cdots + a_{nn}x_{n} &= \sum_{k = 1}^{n}a_{nk}x_{k} = 0.\end{aligned} This can be also represented in matrix form as $A\vec{x} = o$. We can reduce this system to system of $n-1$ equations by substitution. Respectively: $x_{n}=-\frac{1}{a_{nn}} \sum_{k = 1}^{n-1}a_{nk}x_{k}.$ Now, plugging in back to the $j^{\text{th}}$ equation ($1 \leq j \leq n-1$), we get: \begin{aligned} a_{nn}\sum_{k = 1}^{n-1}a_{jk}x_{k} - a_{jn}\sum_{k=1}^{n-1}a_{nk}x_{k} = 0 \\ \sum_{k = 1}^{n-1}\left(a_{nn}a_{jk} - a_{jn}a_{nk} \right )x_{k} = 0\end{aligned}

We are now left with one less equation and one less variable, but with new transformation matrix $B$ such that $b_{jk} = a_{nn}a_{jk} - a_{jn}a_{nk}$. The idea is to prove general formula for evaluating determinant of $n \times n$ matrix by building upon the basis cases which are simple and intuitive.

- 2 years, 9 months ago

Read up the history of the determinant; it is actually quite interesting. In any case, I think determinants arose as a consequence of solving a system of linear equations; if you have a system of $n$ linear equations in the variables $x_1, x_2, \dots, x_n$ and expressed them in terms of $y_1, y_2, \dots, y_n$ then inverting the system naturally gives you this quantity called a determinant. The area/volume/hypervolume story came much later, dating back to Euler and Tartaglia, but formalised by Cayley & Menger. If you want to see how Euler derived his story of volumes in terms of determinants, have a look at Dorrie's book 100 Great Problems in Elementary Mathematics (apologies in advance if I get the name wrong).

- 2 years, 7 months ago

Wonderful Youtube channel with intuitive videos. Here is their video on the determinant.

- 2 years, 7 months ago

Actually, I watched that video and all other from Linear Algebra series. I'm familiar with Grant Sanderson's content and I really enjoy it. However, except for the intuition I mentioned in my note above, the video doesn't offer an answer to my second question, namely, how is the formula derived for $n$ dimensions.

- 2 years, 7 months ago

Oh, okay! No problem.

- 2 years, 7 months ago