Is there a suitable macro (or even style) for responses to referees and editor?

These often entails several points to that need to be addressed. It would be nice to, say, quote the original text (possibly numbered) and the show the response slightly indented, as in

a) Spell foo as bar

We opted to follow the ABC style of ...

b) Include reference to Really, Important and Stuff (2010)

Section 2 has been updated and the discussion ...
  • Maybe use a description list with some counters? Commented Aug 25, 2010 at 18:16
  • 3
    Are you prepared to type in the original report, or do you have the source for that? Commented Aug 25, 2010 at 18:26
  • 1
    Fair question. I think I'd copy it: it may be ascii, or rendered to ascii via pdftotext. Commented Aug 25, 2010 at 18:49
  • You can use the enumitem package to customise the layout of lists. I think that's pretty much everything that you will need; you can use any document class like article. Commented Aug 25, 2010 at 22:29
  • 2
    Excellent question. @Andrew: Including cut-out chunks of pdf from the original might work out to be the best strategy. By coincidence, last week I received a response to referees where the contents of my report were rather mangled. The authors used four text styles, to indicate referees original, their comments, text they cited from their original article, and new text they inserted into their article, which struck me as a good way to go about things. Commented Aug 26, 2010 at 10:07

8 Answers 8


Given the lack of pointers to anything pre-existing, I ended up with the following (and rather simplistic) solution, at least for now:

\newcommand{\pointRaised}[2]{\medskip \hrule \noindent 
               \textsl{{\fontseries{b}\selectfont #1}: #2}} 
\newcommand{\reply}{\noindent \textbf{Reply}:\ }       

The \pointRaised macro takes two arguments. The first is the enumeration and/or location of the issue. This will be set bold. The remainder is the what I chose to quote from the issue raised. This will be set in slanted. The whole block is separated by some vertically space and a horizontal rule.

The \reply simply starts with a bold token followed by the comment.

  • Old answer but I found a nice solution. I use the environment leftbar from the package framed and the quotation environement to display the referee's text from my replies. Commented Mar 1, 2019 at 9:58
  • Can you provide this solution in a full example?
    – alper
    Commented Sep 28, 2023 at 14:06
  • Old question :) I just use the two commands listed in a series of paragraphs. List a question with \pointRaised{somelabel}{more text here} then followed by \reply More text. Commented Sep 28, 2023 at 14:10

For insert comments about the text modifications in a LaTeX file according to referee questions, a good and easy option could be make a version with the todonotes package activated. Small comments can be made at margin and big comments might be placed in line. For example:


\title{A more clever title
\todo{Modified according to referee}}
\author{The author}
{Abstract changed to adapt to format indicated in
guidelines to authors. Text has beeen changed to 
reflect the update of Section 2 and Discussion.}
Really et al. (2010) 
\todo[color=blue!40]{Added citation} 
said some important suff.\lipsum[2]

For a letter with questions & answers, this is essentially the same text structure of a exam, so you can use the exam document class. An example:



\renewcommand{\thequestion}{\alph{question} }



Dear Editor in Chief, \bigskip

Please read below our answer about the \numquestions{}
questions made by  both refereees. I hope that you agree
wih all our comments. \bigskip

Best regards,\bigskip

The author

\question Spell foo as bar
 We opted to follow the ABC style of ... 
\question  Include reference to Really, Important and Stuff (2010)
Section 2 has been updated and the discussion ... 

  • 1
    In my discipline, I'm pretty sure this would annoy the hell out of both editor and referees!
    – cfr
    Commented May 9, 2016 at 0:53
  • 1
    Yes, many of them are quite fussy :) I real life I have never sent colored letters to the editor and galley proofs , but I just pretended show the possibilities, not suggest a gaudy style of reply. At this respect my only suggestion is: contact them exactly as they ask that you contact them. `:)
    – Fran
    Commented May 9, 2016 at 6:29
  • As a journal editor, I do not like the marked up text alone, because I look for the author to tell me for each point I or the referee raised, exactly what the response is. Sometimes a single point will necessitate changes in multiple places in the text, and sometimes nothing will change. With the markup only, there's no way to identify how a specific comment was answered. The "exam" style is a good idea, and a comment/response alone is acceptable.
    – Liam
    Commented Aug 14, 2018 at 20:09

There is a new template for a beautiful authors response to reviewers, that is

  1. easy to write, and
  2. easy to read.

The content can be written like this:

\section{Reviewer 1}
\subsection{Page 4 Line 12}
\RC Lengthy paragraph about something
\AR I do not agree.
Excerpt from the text with latexdiff-flavoured text changes.

Or alternatively, in Markdown:

# Reviewer 1
## Page 4 Line 12
\RC Lengthy paragraph about something
\AR I do not agree.
> Excerpt from the text with latexdiff-flavoured text changes.

And looks like this:

Example author response to review letter

  • Is the typo in the title on purpose?
    – Johannes_B
    Commented May 20, 2017 at 5:48
  • Thanks Johannes_B, I fixed the typo and added even more features to the template.
    – Martin
    Commented May 29, 2017 at 8:21
  • Why everthing is italic?
    – alper
    Commented Sep 28, 2023 at 13:18
  • @alper To indicate that this is written speech of a discussion between two people. This is also to better distinguish from the quoted text of the manuscript, which is the main subject of the whole discussion.
    – Martin
    Commented Sep 28, 2023 at 17:18

I've modified this idea a bit, adding counters to automatically number the responses, and automatic referencing of where you've made relevant changes:


% counters for reviewer points

                  \section*{Reviewer \#\thereviewer:}}
\newcommand{\point}[2]{\stepcounter{point} \bigskip \hrule \medskip \noindent 
               \textsl{{\fontseries{b}\selectfont (\thereviewer.\thepoint) #1} #2}} 
\newcommand{\reply}{\medskip \noindent \textbf{Reply}:\ }   

% use this command in the text where a change addressing a reviewer point has occurred
% and this one to refer to such a location
\newcommand{\revref}{{(p.\ \pageref{rr:rev\thereviewer:\thepoint}, l.\ \lineref{rr:rev\thereviewer:\thepoint})}}


% put this where the reviews are to appear (at the end?)
      {\LARGE \bf Response to Reviews}

To use this, I've got the tex for the paper in, say, paper.tex, and the responses in responses.tex (just the document, no header commands), and in the header of paper.tex:

\reviewresponsestrue  % include them
% \reviewresponsesfalse  % don't include them
\newcommand{\responsefile}{responses.tex}  % name of the review reponses file

Just after the \begin{document}:


And down where I want the responses to appear in the document:

% put in the reviews here

For instance, when I add some text or whatnot to address, say, Reviewer 2's third point, for example:

This point was previously tangentially alluded to by Fred et al (2007). \revpoint{2}{3}

the responses might look like

\point{p.7}{This was previously studied by Fred et al (2007), which should be cited.}

\reply{We have included the citation \revref.}

This is working great; the only unsatisfactory thing about it is that you can't reorder reviewer points or add/subtract any after you've started working or you mess up the whole numbering scheme. It seems preferable to making up unique keys for each point yourself or something. Any other ideas?


This will be a non-LaTeX. As Editor in Chief for a journal, I will add some thoughts based on experience. The lack of templates reflects the fact that there is no real need for such a template. A letter in rresponse to referees comments should probably be made so that it is clear that each comment has been responded to. Such a letter will then probably reflect the review format more than enything else.

I would argue that the most important aspect of such letters are clearly non-(La)TeX issues summarized by 1. clarity 2. sincerity 3. cordiality

As is stated by @dank normal structural features of (La)TeX is enough (technically) to make a successful response letter.

  • 1
    I agree, I really liked the answer by @petrelharp but even that was too much for (even though the cross-referenec idea is cool). Currently working a revision and letter, and still just using the basic two macros for minimal, yet consistent and automated, markup. Commented Jan 8, 2013 at 21:06
  • 1
    @Dirk Eddelbuettel, thanks for your comment. I realized when reading it that it may be useful to provide some form of review report-review response package so that if one uses the package for a review the reviewer can use the .tex file and apckage for a structured response. It would not be a template but more of different functions for comment-repsonse. The reviewer would then be the one deciding on the structure and the responder will simply follow. Could be interesting! Commented Jan 8, 2013 at 21:15
  • I definately agree with this; but I've found the automatic line referencing to be really useful, since otherwise, inevitably, I will re-edit something and forget to update the response.
    – petrelharp
    Commented Feb 21, 2013 at 17:06

Just use itemize or enumerate. Put the suggestion (or, better, a one-line summary of it) in bold or italic at the start of each item.

Although it is useful to use latex or tex (in case there is mathematics involved), the formatting should not matter much to the editor or the reviewers, so long as things are clear. Don't waste time worrying about formatting.

Instead, spend your time documenting the changes to your document (or reasons for not changing). I normally quote line numbers in the revised version, so that reviewers can check quickly the changes that have been made, to see if they judge them sufficient. Few reviewers are going to read the whole text again, with the original one at its side, to see the changes.

The reviewers have done a lot of work for you, and few systems credit that work. (Deans and bosses of all kinds want to count beans, and providing careful reviews counts as, well, as a first approximation, zero beans.) You should give your reviewers a break, in the second round. Tell them what you've changed (and where in the text), and why you refuse to change other things. If you have some minimal formatting (even a blank line between items), that's sufficient.

  • 3
    Just for closure, I do not consider Don't do it to be a helpful response to a detailed how do I question. But thanks anyway. Commented Aug 29, 2010 at 12:34
  • @DirkEddelbuettel I don't consider this to be a 'Don't do it'. It is definitely a 'do it'. You may disagree about the suggested formatting, but you did ask for suggestions. This is a pretty standard way of doing it, I think. This is the best of the answers I've read so far. At least, maybe it is different in your discipline. In mine, this is the best, I think.
    – cfr
    Commented May 9, 2016 at 0:57

I created this template if anybody needs it: https://sites.google.com/site/andreaballatore/misc

  • 3
    Page not found! Please re-build your link!
    – Qaher
    Commented Sep 7, 2017 at 15:40

As I like the simplicity of Dirk Eddelbuettel's answer (petrelharp's solution was a bit too complex for my taste), but don't like keeping track of point numbers (I got a lot of comments from my reviewers), here's my take using a point counter and an environment to reset the counter for new reviewers.

% response_macros.tex


\newcommand{\point}[1]{\medskip \noindent
               \textsl{{\fontseries{b}\selectfont Q\thepointcounter}.
                 \stepcounter{pointcounter} #1}}
\newcommand{\reply}{\medskip \noindent \textbf{Answer}.\ }

And you use it as...


\section*{Reviewer 1}

\point{Could you fix this?}
\reply Sure, I can.

\point{And that?}
\reply Why not?


\section*{Reviewer 2}


\reply Thanks.

  • Thanks for that, I may use this next time as it is such as obvious improvement. IIRC when I asked originally, I may have had a reviewer with a numbered list. If it is just bullet points this automagic counter use is clearly better. Commented Mar 5, 2015 at 17:32

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .