Equation refuses to move leftward, past the first line

I have a really long multiline equation, inside an align environment, which is inside a subequation environment.

Each line of the equation is just a tiny bit too long for both the equation and the equation line number to fit into one line. So, I would like to move the entire multiline equation leftward a little bit, further left than the left edge of the text. However, the equation would only ignore the left margin and move further left for the first line. Any other line in the equation is instead aligned with the left edge of the text.

How should we tell LaTeX to move the entire equation leftwards ignoring the left margin?

Specifically, neither \hspace{-5in} nor \hspace*{-5in} would work.

Below are the first two lines of the equation: (the original equation is 10+ lines long, so)

\documentclass[11pt, oneside]{article}
\usepackage{geometry}
\geometry{letterpaper}
\usepackage[utf8]{inputenc}
\usepackage[T1]{fontenc}
\usepackage[english]{babel}
\usepackage{amsmath}

\begin{document}

\begin{subequations}
\begin{align}
\hspace*{-20cm}
\langle\tilde{\Theta}(\mathbf{l_1})...\tilde{\Theta}{(\mathbf{l_4})}\rangle&=\langle
\Theta(\mathbf{l_1})
\Theta(\mathbf{l_2})
\Theta(\mathbf{l_3})
\Theta(\mathbf{l_4})\rangle\\
+(
\iint\frac{d^2\mathbf{l_1'}d^2\mathbf{l_2'}}{(2\pi)^4}&\langle
\Theta(\mathbf{l_3})
\Theta(\mathbf{l_4})
\Theta(\mathbf{l_1'})
\Theta(\mathbf{l_2'})
\phi(\mathbf{l_1}-\mathbf{l_1'})
\phi(\mathbf{l_2}-\mathbf{l_2'})\rangle
[(\mathbf{l_1}-\mathbf{l_1'})\cdot \mathbf{l_1'}]
[(\mathbf{l_2}-\mathbf{l_2'})\cdot \mathbf{l_2'}]
\end{align}
\end{subequations}

\end{document}

• Please, add an example of code. Jul 14, 2017 at 8:18
• Assuming your equations contain several =, +, and - symbols, I'd look into reducing the spacing around these symbols ( by reducing the values of thickmuskip and medmuskip) before considering letting the equations be wider than the text block. Are you at liberty to post the contents of the align environment? If so, please edit your posting to show the multiline equation, along with information such as the width of the text block, the font size, and the font itself.
– Mico
Jul 14, 2017 at 8:20
• Yes -- code added. I was just worrying that maybe it would compile differently on your sides and you would not seeing this effect. Jul 14, 2017 at 15:10
• your code is not compilable. what is \vt?. please provide complete small document, that we can see, what you have in preamble (page layout set). Jul 14, 2017 at 15:31
• please, extend your code sniped to complete small document, that we can see page layout set. If this is one equation, then try to use multline instead of align. Check equation again, it seems that has some math errors. Jul 14, 2017 at 15:43

Locally you can enlarged text width with \adjustwidth environment from changepage package and with it move equation to the left with centered in the wide environment:

    \documentclass[11pt, oneside]{article}
\usepackage[showframe]{geometry}
\geometry{letterpaper}
\usepackage[utf8]{inputenc}
\usepackage[T1]{fontenc}
\usepackage[english]{babel}
\usepackage{amsmath}

\usepackage[strict]{changepage}

\begin{document}

\begin{subequations}
\begin{align}
\langle\tilde{\Theta}(\mathbf{l_1})...\tilde{\Theta}{(\mathbf{l_4})}\rangle&=\langle
\Theta(\mathbf{l_1})
\Theta(\mathbf{l_2})
\Theta(\mathbf{l_3})
\Theta(\mathbf{l_4})\rangle\\
+(
\iint\frac{d^2\mathbf{l_1'}d^2\mathbf{l_2'}}{(2\pi)^4}&\langle
\Theta(\mathbf{l_3})
\Theta(\mathbf{l_4})
\Theta(\mathbf{l_1'})
\Theta(\mathbf{l_2'})
\phi(\mathbf{l_1}-\mathbf{l_1'})
\phi(\mathbf{l_2}-\mathbf{l_2'})\rangle
[(\mathbf{l_1}-\mathbf{l_1'})\cdot \mathbf{l_1'}]
[(\mathbf{l_2}-\mathbf{l_2'})\cdot \mathbf{l_2'}]
\end{align}
\end{subequations}

$$\begin{multlined} \langle\tilde{\Theta}(\mathbf{l_1}) \dots \tilde{\Theta}{(\mathbf{l_4})}\rangle = \langle \Theta(\mathbf{l_1})\Theta(\mathbf{l_2})\Theta(\mathbf{l_3})\Theta(\mathbf{l_4}) \rangle\\ +( \iint\frac{d^2\mathbf{l_1'}d^2\mathbf{l_2'}}{(2\pi)^4} \langle \Theta(\mathbf{l_3})\Theta(\mathbf{l_4})\Theta(\mathbf{l_1'})\Theta(\mathbf{l_2'}) \phi(\mathbf{l_1}-\mathbf{l_1'})\phi(\mathbf{l_2}-\mathbf{l_2'}) \rangle [(\mathbf{l_1}-\mathbf{l_1'})\cdot \mathbf{l_1'}] [(\mathbf{l_2}-\mathbf{l_2'})\cdot \mathbf{l_2'}] \end{multlined}$$
\end{document}

Note, so far I didn't see that part of one equations is numbered. See, if the use multlined is better suit to what you like to have.

\documentclass[11pt, oneside]{article}
\usepackage[showframe]{geometry}
\geometry{letterpaper}
\usepackage[utf8]{inputenc}
\usepackage[T1]{fontenc}
\usepackage[english]{babel}
\usepackage{mathtools}

\usepackage[strict]{changepage}

\begin{document}

\begin{subequations}
\begin{align}
\langle\tilde{\Theta}(\mathbf{l_1})...\tilde{\Theta}{(\mathbf{l_4})}\rangle&=\langle
\Theta(\mathbf{l_1})
\Theta(\mathbf{l_2})
\Theta(\mathbf{l_3})
\Theta(\mathbf{l_4})\rangle\\
+(
\iint\frac{d^2\mathbf{l_1'}d^2\mathbf{l_2'}}{(2\pi)^4}&\langle
\Theta(\mathbf{l_3})
\Theta(\mathbf{l_4})
\Theta(\mathbf{l_1'})
\Theta(\mathbf{l_2'})
\phi(\mathbf{l_1}-\mathbf{l_1'})
\phi(\mathbf{l_2}-\mathbf{l_2'})\rangle
[(\mathbf{l_1}-\mathbf{l_1'})\cdot \mathbf{l_1'}]
[(\mathbf{l_2}-\mathbf{l_2'})\cdot \mathbf{l_2'}]
\end{align}
\end{subequations}

$$\begin{multlined} \langle \tilde{\Theta}(\mathbf{l_1}) \dots \tilde{\Theta}(\mathbf{l_4}) \rangle = \langle \Theta(\mathbf{l_1})\Theta(\mathbf{l_2})\Theta(\mathbf{l_3})\Theta(\mathbf{l_4}) \rangle\\ +( \iint\frac{d^2\mathbf{l_1'}d^2\mathbf{l_2'}}{(2\pi)^4} \langle \Theta(\mathbf{l_3})\Theta(\mathbf{l_4})\Theta(\mathbf{l_1'})\Theta(\mathbf{l_2'}) \phi(\mathbf{l_1}-\mathbf{l_1'})\phi(\mathbf{l_2}-\mathbf{l_2'}) \rangle [(\mathbf{l_1}-\mathbf{l_1'})\cdot \mathbf{l_1'}] [(\mathbf{l_2}-\mathbf{l_2'})\cdot \mathbf{l_2'}] \end{multlined}$$

Try with \begin{flalign}...\end{flalign} which comes with the package amsmath.sty
• The OP mentions: "Each line of the equation is just a tiny bit too long for both the equation and the equation line number to fit into one line." How would flalign help in this situation?