# Why is my proof not working with a matrix?

I was writing up a proof on a LaTex article, which required me to utilise a number of matrices in it.

However, I quickly discovered that when I hit \end{proof}, at the end of my proof, it would not register as the end of my proof, or even recognise the proof command, thus not including the end proof box, and I quickly noticed that even the beginning proof. had disappeared.

I experimented with other options: Putting the begin{proof} and end{proof} on either side of a paragraph with no matrices seemed to work fine. But the moment I added a matrix using the following code, I ran into a problem:

4. Two columns/rows of the matrix are identical. This is simply a property of a matrix.
\begin{Proof}
Let us say we have an n x n matrix A, shown below:
$\begin{bmatrix} a_{11} & a_{12} & a_{13} & ... & a_{1n}\\ a_{21} & a_{22}&... &... &a_{2n} \\ a_{i1} & ...&... &... &a_{in} \\ a_{n1} & ...&... &... &a_{nn} \\ \end {bmatrix}\\$
\end{Proof}


It shows up like this: With no proof.

It is not a problem with my packages, I have used proofs successfully throughout the document, and here is my preamble:

\documentclass{article}
\usepackage{amsthm}
\usepackage{amsmath}
\usepackage{amssymb}
\usepackage[margin=1.25in]{geometry}
\usepackage[shortlabels]{enumitem}
\usepackage{graphicx}
\usepackage{fancyhdr}
\renewcommand{\footrulewidth}{1pt}
\fancyhfoffset{0.2cm}
\pagestyle{fancy}
\rfoot{redacted}
\newtheorem*{theorem}{\sc{Theorem}}
\newtheorem*{definition}{\sc{Definition}}
\newtheorem*{proposition}{\sc{Proposition}}
\newtheorem*{corollary}{\sc{Corollary}}
\newtheorem*{claim}{\sc{Claim}}
\newtheorem*{properties}{\sc{Properties}}
\newtheorem*{remark}{\sc{Remark}}
\DeclareMathOperator{\N}{\mathbb{N}}
\DeclareMathOperator{\Z}{\mathbb{Z}}
\DeclareMathOperator{\Q}{\mathbb{Q}}
\DeclareMathOperator{\R}{\mathbb{R}}
\DeclareMathOperator{\C}{\mathbb{C}}


Any ideas why this is happening and what I can do to fix it?

• How or where is the Proof environment defined? – Mico Jun 26 at 6:40
• It is defined as I have used it on multiple occasions throughout the document and it has worked. Only difference is the matrix – global05 Jun 26 at 6:41
• Incidentally, \sc is a "switch", i.e., it doesn't take an argument. You should write either \textsc{...} or {\scshape ...}. – Mico Jun 26 at 6:42
• I see: You have a typo in your code: \begin{Proof} should be \begin{proof}. Also, all 22 [!] instances of $ inside the bmatrix environment have to be deleted. Once you fix these issues, the end-of-proof symbol shows up. – Mico Jun 26 at 6:43 ## 1 Answer You have a basic typo in your code: The amsthm package defines an environment called proof, but not one called Proof. If you use \begin{Proof} and \end{Proof}, and if you choose to ignore plenty of LaTeX warning messages, you'll eventually obtain the following output: Indeed, no "Proof" label up front, and no end-of-proof symbol at the end. Happily, once you change \begin{Proof} and \end{Proof} to \begin{proof} and \end{proof}, respectively, you'll get the following output: Moral of the story: Never, ever ignore LaTeX's warning messages. (PS To make your code compile, I also had to delete all 22 instances of $ inside the bmatrix environment.)

\documentclass{article}
\usepackage{amsthm,amsmath}
\usepackage[margin=1.25in]{geometry}
\usepackage[shortlabels]{enumitem}

\begin{document}
\begin{enumerate} \setcounter{enumi}{3} % just for this example
\item Two columns/rows of the matrix are identical. This is simply a property of a matrix.
\begin{Proof} % <-- typo
Let us say we have an $n \times n$ matrix $A$, shown below:
$\begin{bmatrix} a_{11} & a_{12} & a_{13} & ... & a_{1n} \\ a_{21} & a_{22} &... & ... & a_{2n} \\ a_{i1} & ... &... & ... & a_{in} \\ a_{n1} & ... &... & ... & a_{nn} \end{bmatrix} \qedhere % optional$
\end{Proof} % <-- typo
\end{enumerate}
\end{document}

• You marked the typos but I think you forgot to correct them ;-) – campa Jun 26 at 7:20
• Ah, the attention to detail. This worked. – global05 Jun 26 at 7:43
• @campa - Fixing the very last few details is left as an exercise to the readers, right? (I did eliminate all 2*11=22 [!] spurious \$ symbols…) – Mico Jun 26 at 8:59