If you decide to collaborate with someone, whether as author or editor (where here I include merely making suggestions), choose your collaborator(s) carefully, with due attention to personal compatibility and relative expertise. At each stage, all parties should make every effort to ensure that what they forward to another party is as good as they can make it: there is no point in wasting time on silly mistakes the original author should have spotted and corrected.

Various schemes are available for facilitating the process; some are incorporated as features in certain software products. Here I will offer comments on a few possibilities. It will be understood that references to author and editor reflect their roles in a particular situation: the roles may be reversed many times in the course of the collaboration. Note that in some cases author and editor may be one and the same person!

• Tracked Changes

  I am not a fan of Tracked Changes. With few exceptions, most documents in my experience require numerous changes, and the display of Tracked Changes can appear very messy. A lot of these changes would be unnecessary if the author had thought more about their writing in the first place. Many of the errors would almost certainly have been obvious if the author had gone to the trouble of reading what they had written. Beware of the author who only uses Tracked Changes: it’s possible they’re just being lazy. They expect someone else to do the hard work and then content themselves with a simple Accept or Reject to effect the suggested changes (or not). They aren’t interested in discussion, and there is little opportunity for them to learn from the exercise and become a better writer. It should also be borne in mind that the editorial change suggested could itself harbor some error that is too easily missed in the course of injudicious use of the Accept button.

• Margin Comments

  Margin Comments at least offer the possibility of explaining why the editor has made the suggested alterations. This can provide a learning experience for the author. Margin Comments also admit of suggestions which do not consist simply of alternative forms of expression. Structural change would be an example. This modality also allows other sorts of statements (such as approval or disapproval, or references elsewhere), as distinct from strings for direct insertion or replacement. These other sorts of statements are logically separate from entries in a Tracked Changes log and so are more appropriately located elsewhere. Like Tracked Changes, though, Margin Comments can be messy in appearance, with pointer lines criss-crossing all over the place and possibly some ambiguity in the case of remarks concerning nested strings.

• Footnotes

  Footnotes offer a much cleaner appearance. In a document that contains no footnotes of the usual sort, comment footnotes are easily spotted. If necessary, editorial remarks can be distinguished from other footnotes by a suitable choice of font, font style or color. Footnotes can be amplified or deleted easily. They may alter pagination until they are removed, but implementing changes of any sort can do this anyway. I often favor the Footnote approach when I’m collaborating with myself. Rather than dwell overly long on a particular piece of text, I might simply add a footnote indicating where more work needs to be done and what my concern is.

• Inserted Annotations

  The idea here is simply to insert a new paragraph after a passage that requires comment. This paragraph should be distinguished from regular text in some way; eg, by italicization, color or a predefined style. Further comments can be added by either party. When the matter has been dealt with, the interpolated paragraph can just be deleted. This is a simple method, usable with any writing software and easily understood by people having little expertise with word processors. I use it a lot.

• Markup

  Various Text Markup schemes (akin to HTML coding) are available and may be worth considering by the more technically inclined in certain situations.

Whatever scheme you use, make sure you test the compatibility of software implementations of relevant features on all devices involved, before you commit yourself to a collaborative project. DOCX now seems to be the preferred document format, even when collaborating with people who use a different word processor and/or operating system from your own. Also, when working on text, it’s better to work with a text-only document, as the handling of graphics elements can vary with the environment. Leave layout for another day (and perhaps another person!).

Finally, don’t forget that handwritten comments on printouts may suit certain participants better than more technical methods.

Sometimes an author wishes to apprise the reader of a character’s unspoken thoughts in the course of spoken dialogue. (Some commentators refer to this as “articulated thought”.) In the case of a script for stage or screen, this is usually handled by an aside, and there is no confusion about which character is involved and which parts of the text represent thoughts and which represent utterances to other characters. In the case of standard prose, however, some other device is required to effect the distinction.

One approach would be to enclose the thought in quotation marks and have it preceded or followed by some explicit form of words (sometimes called a “speech tag”) that indicates its status as unspoken thought. Simple examples of this approach are:–

In the first example, we get to the end of the sentence before we realize Mary is not actually replying aloud to Bob’s latest utterance. In the second example, the structure is too stilted for most purposes, especially if the second speech tag precedes the spoken part, which it really should for clarity. And neither format lends itself to much repetition, especially in the course of a single block of dialogue.

Another stratagem is to forego the use of quotation marks when delineating unspoken thoughts and cast them in italics instead. This has the advantage that subsequent occurrences might be able to dispense with explicit descriptions of the nature of the italicized remarks. However, italicized text is often used for other purposes (such as titles of books and movies, or designating statements of significance), in which case confusion can arise.

Yet another approach is to avoid the use of quotation marks or italics and simply rely on speech tags to identify the status of unspoken remarks. To my mind, it doesn’t have quite the same force as the other methods. It relegates the text to the same status as action in the main narrative, whereas really it represents privileged information. It presents difficulties when the unspoken thought continues over more than one sentence: how many speech tags are needed and where does one place them? I don’t think this tactic is sustainable in the course of a dialogue. The distinction between unspoken and spoken text is not so clear-cut as one might think. Furthermore, the reliance on speech tags in this context is taxing on the writer and tiresome for the reader.

I find it remarkable that in these days of computerized manuscript creation, so little is made of the multiple formatting styles available when it comes to ordinary narrative text. Unspoken thought presents itself as a prime candidate for creative treatment.

I therefore suggest adoption of a suitably contrived formatting style to be applied to unspoken thought. Elements of such a style might include a different font and/or a different color from those used in the rest of the text. This admits of formatting unique to each character. The use of this technique would be similar to that relating to the italicization technique described above, but it might even be possible to dispense with explicit descriptions entirely, even with the first appearance of unspoken thought in the manuscript, especially if the content of the thought is patently something the character would not dare to utter aloud to other parties in the dialogue. It would be the exceptional case where an astute reader would not quickly apprehend the significance of the stylized text. Given that parentheses never occur in spoken text, enclosing an unspoken thought in parentheses might be a good idea if the special format is not sufficiently different from that of the rest of the narrative (if they still use the same font, for example). I would still enclose the unspoken thought in quotation marks to indicate we are reporting the exact wording of the character’s “internal speech”. Note that use of different fonts in the same paragraph can affect line spacing, so a fixed line height may be appropriate in such a paragraph.

I’ll give a few examples for your consideration. All employ a speech tag for the first occurrence of an unspoken thought and then use parentheses and special formatting for each subsequent occurrence. The first simply involves setting the entire parenthesized expression in a different color. One should be mindful not to choose colors that may prove problematic for color-blind readers.

The second also involves no change of font but sets the entire parenthesized expression in a shade of grey (or a lighter shade of whatever color is used for the main narrative text). This technique connotes a sotto voce mode of speech which seems to be appropriate.

My third example uses a different font (serif rather than sans serif) and font weight, the combined effect of which is to give more prominence to the unspoken thoughts. As far as Mary is concerned, the unspoken thoughts may well be more prominent in her mind anyway.

Admittedly, such special formatting of unspoken thoughts can appear somewhat peculiar at first glance. But readers have proven themselves quite adaptable in the face of all sorts of peculiarities foisted upon them by writers. The present scheme affords more intimate access to a character’s mind than more conventional approaches. It gives the writer more scope in narration — something a reader may actually appreciate, to the extent they are encouraged to disregard any discomfort felt on initial acquaintance.

With the above thoughts as guidelines, you might like to perform your own experiments and settle on a technique that appeals to you. Then be consistent!