Foundations — Types of matrices — row, column, square, diagonal, identity, zero, symmetric, skew-symmetric

Before you can say "this matrix is symmetric" or "that one is skew-symmetric", you must be fluent in the tiny alphabet the parent note quietly assumes. This page builds every one of those pieces from nothing. We go slowly and in the exact order the ideas depend on each other.

---

1. What is a matrix, really?

Picture a spreadsheet, a chessboard, or the seats in a cinema — anything laid out in a neat grid.

Look at the figure. The numbers going across (left to right) form a row. The numbers going down (top to bottom) form a column. That is the whole idea — everything else is vocabulary describing this grid.

---

2. Rows, columns, and the symbol $a_{ij}$

To talk about one specific number inside the grid, we need an address — like "row 2, column 3".

In the figure, the highlighted cell is $a_{23}$: go down to row 2, then across to column 3. That single grey cell is the value that name points to.

The whole point of the topic — spotting types — is really just spotting patterns in the $a_{ij}$. "Symmetric" will turn out to mean $a_{ij}=a_{ji}$, i.e. swapping the two indices changes nothing. We could not even write that condition without this symbol.

---

3. Order (size) of a matrix: $m\times n$

This one number-pair, $m\times n$, is what the parent note uses to define the first three types:

• Row matrix = order $1\times n$ (exactly one row).
• Column matrix = order $m\times 1$ (exactly one column).
• Square matrix = order $n\times n$ ($m=n$: rows equal columns).

So "types by dimension" is literally just which numbers $m$ and $n$ are allowed to be.

---

4. The principal diagonal

Once a matrix is square ($n\times n$), one special line of entries gets a name.

The figure highlights the diagonal in yellow. Notice the address rule: every yellow cell has matching indices ($a_{11},a_{22},a_{33}$). Every cell off the diagonal has $i\ne j$.

Why does this line matter so much? Because almost every special square-matrix type is defined by what lives on the diagonal versus off it:

• Diagonal matrix: everything off the diagonal is $0$ (only the yellow cells may be non-zero).
• Identity matrix: diagonal is all $1$s, off-diagonal all $0$s.
• Skew-symmetric matrix: the diagonal is forced to be all $0$s (proved in the parent note).

The diagonal also splits the grid into an upper triangle (cells above it, $j>i$) and a lower triangle (cells below it, $i>j$) — this is the mirror line for symmetry.

---

5. The transpose $A^T$ — the mirror flip

The last two types (symmetric, skew-symmetric) are defined using $A^T$, so we must build it now.

Look at the figure. Reflecting the grid across the principal diagonal (the dashed yellow line) sends each entry to its "mirror" position. The diagonal entries sit on the mirror line, so they never move.

Now the two symmetry types read cleanly:

• Symmetric: $A^T=A$ — the mirror flip changes nothing, so $a_{ij}=a_{ji}$.
• Skew-symmetric: $A^T=-A$ — the mirror flip negates everything, so $a_{ij}=-a_{ji}$.

For a deeper treatment of the flip itself, see Matrix Transpose.

---

6. Two special numbers: $0$ and $1$ (for matrices)

The parent note says the zero matrix "behaves like 0" and the identity "behaves like 1". Here is the plain meaning.

These "do-nothing" matrices are the anchors for the whole theory — the identity is what an Inverse of a Matrix must produce ($A^{-1}A=I$), and it reappears in Eigenvalues and Eigenvectors and Determinants.

---

How these foundations feed the topic

Every arrow is a dependency: you cannot understand "symmetric" (bottom) without "transpose" (middle) without "$a_{ij}$" (top). Build upward. The finished topic is Types of Matrices, and it opens the door to Determinants and Diagonalization.

---

Equipment checklist

Try to answer each before revealing. If any stump you, re-read that section above.