writing peeves: use first person and active voice or the red pen of wrath will rain a rewrite upon thee

In my undergraduate classes, students always express surprise when I say that I expect them to use first person when appropriate in writing, including field notes and the methods section of research papers.   I explain that disciplines and instructors have different norms about voice, but my preference is for conveying that an actor is responsible for an action.  My undergrads usually embrace first person after this short explanation.

However, some writers find breaking the passive voice habit a difficult endeavor.  Manuscripts and publications reveal that some colleagues rely on passive space-wasters such as “There is/are…” rather than explicitly identifying actors or assigning responsibility for actions.  When encountering such fillers in manuscripts and publications, I have the same visceral reaction as a craftsperson spotting a poorly executed part in a colleague’s handiwork – an immediate urge to transform the offending sentence into a tighter, more informative one.*  It seems I’m not the only one.

One journal editor shares similar rationales for using first rather than third person:

First person pronouns

This is one area in which I am not conservative. In the olden days we were taught to write in the third person – “the author found” – and in passive voice – “it was found that….” This is really a rhetorical device to convince readers to believe you and not ask embarrassing questions. The results were discovered on stone tablets on the mystic mountain.  No human beings were involved in the research. Of course it’s correct! It’s SCIENCE.

In the last thirty years or so, people in the philosophy of science have been dealing with the fact that human beings actually are involved in research, that they have interests and biases, and that the best way to develop some kind of shareable knowledge, otherwise known as “science” or “objectivity,” is to admit these biases and try to control them rather than pretending they don’t exist. Post-modernists and phenomenologists seem to have given up on shareable knowledge.

At JSSW we still look for it, we just realize that it’s harder to attain than we used to think.

So, it’s fine with me if you want to speak in the first person. But if you’re of the old school, I won’t try to change you. I’ll settle for trying to make you uncomfortable.

William H. Starbuck shares his tips for how to avoid first person, without resorting to third person:

Avoid first-person and second-person pronouns.

First-person pronouns do sometimes serve well, but you should recognize the disadvantages of saying I, we, our or you. When you refer explicitly to yourself, you remind your readers that they are reading your ideas, not their ideas; and consequently you make it more difficult for your readers to adopt your ideas as their own. Thus, taking credit for your ideas makes your writing less persuasive and gives your ideas less impact. For example,

In this paper we argue that focusing on –> Focusing on

It is my assumption that the support problems stated are common to any bank service bureau. –> Such support problems afflict any bank service bureau.

I will concentrate my efforts on the coordination mechanisms between departments. –> This research focuses on coordination between departments.

We suggest, furthermore, that assessing –> Furthermore, assessing

In presenting our typology, we will therefore give concrete examples –> Concrete examples illustrate each type.

For similar reasons, do not talk about the reader or the author.

* I would much rather follow the writer’s argument than break out the red pen of wrath.

Written by katherinechen

October 3, 2013 at 12:34 pm

Posted in pet peeve, uncategorized

Tagged with

9 Responses

Subscribe to comments with RSS.

  1. Agreed. I think that there’s something to the sociology of knowledge of point, in the sense that we are most likely to fall back on the impersonal, precisely when conveying those parts of the research that we most anxiously want to portray as objective, because they actually are made up of a bunch of subjective decisions. In quantitative data analysis (where you are likely to find lots of culprits) this is particularly true of paper sections dealing with variable selection, variable coding/recoding, scale construction, and model specification. I used to be a third person addict back in the early grad school days, but little by little I broke the habit and now enforce the rule on graduate student papers that I’m asked to provide feedback for. At the beginning I found it awkward (to write methods sections in the active voice, but now I’m pretty good at catching myself). Maybe other people find it awkward too. (Here’s an entire methods/coding/results section written with lots of sentences that begin with “I…”, starting on section 4 (data and variables).



    October 3, 2013 at 1:23 pm

  2. I too am a big fan of using the first person, precisely for the reasons mentioned above. However, I wonder to what extent native English speaking editors and reviewers realize and take into account how difficult it is for non-native English speaking authors (such as myself) to “get it right” regarding such language preferences. For instance, I submitted a manuscript yesterday to a journal that, coincidentally, specifically stipulates that authors should not use the first person. For my manuscript, this would entail extensive revisions. I’m reluctant to make such changes, given that the manuscript has already been language checked by a professional service at substantial cost. More generally, I occasionally notice that presumably native English speaking reviewers can get agitated over subtle wording choices in a way that for me is impossible to predict (and therefore also to prevent). So, my appeal would be: it’s fine to engage authors about their style but be mindful that he/she may have no way of knowing better.



    October 3, 2013 at 2:15 pm

  3. This is something of an issue with interdisciplinary work. If memory serves, my wife once submitted to an interdisciplinary journal that did APA style and the reviewers thought her use of first person to describe the methods, etc, was “unscientific.” Of course as sociologists we think phrases like “the dependent variable was measured after the treatment” are scientisitic not scientific insofar as they reify the problem of observation, elide questions like whether work was done by the PI vs RAs, and generally make you sound like a pompous windbag pretending to be central casting Dr Labcoat.



    October 3, 2013 at 4:08 pm

  4. I have certain preferences regarding pronouns, but I don’t think editing should consist of substituting your preferences for someone else’s preferences. I think this has occurred in journal editing, which has resulted in a great deal of delay, and an increase in the number of revise-and-resubmit cycles. For a variety of reasons, people may have different preferences than you, and it might be more productive to let people write however they want and focus only on the larger question–is the content worthwhile.
    Note that I’m not saying there’s nothing to be gained by nudging people in a certain direction, and clear writing is certainly important. But I think the place to emphasize your preferences is in the “Instructions to Authors” section.


    Chris M

    October 4, 2013 at 8:21 pm

  5. I actually disagree that passive voice should be avoided. Sure, active sentences can sound nicer, but passive is a useful narrative tool for sociology when we discuss structures and patterns that may in a sense be ‘actorless’ or where we wish downplay agency.

    It’s a waste for the reviewer to spend time correcting grammar and spelling as those sentences may not make it past a revision. It’s no wonder that some journals have slow turnarounds if the reviewers are also trying to act as copy editors.



    October 4, 2013 at 11:08 pm

  6. Reviewers shouldn’t waste their time correcting grammar, I agree, but a good managing editor can help an author get rid of this stuff after they accept the paper. In some cases you may want to retain passive, but most definitely you should avoid it when describing how you constructed variables.


    brayden king

    October 4, 2013 at 11:16 pm

  7. Don’t you mean that “time should not be wasted by reviewers correcting grammar and spelling?”



    October 5, 2013 at 11:56 am

  8. I write all my papers in the subjunctive mood; Would that journal editors were happier with that choice!


    Grammar N.E.R.D.

    October 5, 2013 at 1:29 pm

  9. I have a different approach than most of the commenters here, and the original post. In my Methods sections, I move back and forth between active and passive sentences (or more generally — because this doesn’t always map onto active vs. passive construction, although it often does — sentences that highlight me as the decision-maker vs. sentences that grammatically disguise the decision-maker in order to emphasize the decision).

    My general principle is that decisions that I made that I expect to be controversial, or that are part of the contribution of the paper, or that I otherwise want to implicitly flag for the reader as a *decision* for them to think about, should be grammatically constructed to emphasize that a decision was made (heh). But conversely, less noteworthy decisions should not be so flagged, because long series of sentences that go, “I do this, then I do that” don’t make for great reading and, to my mind, can underemphasize the outcomes of the choices.

    As an example, here’s two sentences from the paper I’m doing the final edits on today.

    To minimize the role played by assumptions about unobserved parameters and maximize the role played by real data in the selection of simulations, I generate many candidate latent subpopulation models and keep only those whose aggregate parameters are consistent with the life tables of known cohorts.

    The mortality hazards are calculated analytically, in instantaneous time.

    The first sentence uses the single I-clause to highlight that I made an important choice about the analytic strategy here. Readers should consider whether this strategy was a good one.

    The second sentence is passive. Readers should know the fact it expresses, and in fact it is a choice I made, but it’s not one where I’m suggesting that readers ask themselves, “Hmm, was she right to do it this way?”

    My papers are full of combinations like this. I definitely don’t claim to be the greatest writer, but to me it makes sense to draw on the full range of grammatical possibility in signaling to the reader about where to focus. If others here think that this is wrong and that all of authors’ analytic choices should be presented in active voice, I’d be interested to know that (seriously — I really do want to improve my academic writing).



    October 5, 2013 at 3:48 pm

Comments are closed.

%d bloggers like this: