We Know the Answer!

By Sean Carroll | May 22, 2007 2:01 pm

Chad Orzel is wondering about the origin of some irritating habits in science writing. His first point puts the finger right on the issue:

Myth 1: First-person pronouns are forbidden in scientific writing. I have no idea where students get the idea that all scientific writing needs to be in the passive voice, but probably three quarters of the papers I get contain sentences in which the syntax has been horribly mangled in order to avoid writing in the first person.

It’s not exactly right to call this a “myth”; as Andre from Biocurious points out in comments, the injuction to use the passive voice is often stated quite explicitly. There’s even an endlessly amusing step-by-step instruction guide for converting your text from active to passive voice. What would Strunk and White say?

The same goes for using “we” rather than “I,” even if you’re the only person writing. There are also guides that make this rule perfectly explicit. The refrain in this one is:

Write in the third person (“The aquifer covers 1000 square kilometers”) or the first person plural (“We see from this equation that acceleration is proportional to force”). Avoid using “I” statements.

Interestingly, these habits did not just emerge organically as scientific communication evolved — they were, if you like, designed. I learned this from a talk given by Evelyn Fox Keller some years ago, which unfortunately I’ve never been able to find in print. It goes back to the earliest days of the scientific revolution, when Francis Bacon and others were musing on how this new kind of approach to learning about the world should be carried out. Bacon decided that it was crucially important to emphasize the objectivity of the scientific process; as much as possible, the individual idiosyncratic humanity of the scientists was to be purged from scientific discourse, making the results seem as inevitable as possible.

To this end, Bacon was quite programmatic, suggesting a list of ways to remove the taint of individuality from the scientific literature. Passive voice was encouraged, and it was (apparently, if Keller was right and I’m remembering correctly) Bacon who first insisted that we write “we will show” in the abstracts of our single-author papers.

It always seemed a little unnatural to me, and when it came time to write a single-author paper (which I tend not to do, since collaborating is much more fun) I went with the first-person singular. I decided that if it was good enough for Sidney Coleman, it should be good enough for me.

Keller has a more well-known discussion of the rhetoric of Francis Bacon, reprinted in Reflections on Gender and Science. Here she examines Bacon’s personification of the figure of Nature, specifically with regard to gender roles. Bacon was one of the first to introduce the metaphor of Nature as a woman to be seduced/conquered. Sometimes the imagery is gentle, sometimes less so; here are some representative quotes from Bacon to give the gist.

“Let us establish a chaste and lawful marriage between Mind and Nature.”

“My dear, dear boy, what I plan for you is to unite you with things themselves in a chaste, holy, and legal wedlock. And from this association you will insure an increase beyond all the hopes and prayers of ordinary marriages, to wit, a blessed race of Heroes and Supermen.”

“I am come in very truth leading you to Nature with all her children to bind her to your service and make her your slave.”

“I invite all such to join themselves, as true sons of knowledge, with me, that passing by the outer courts of nature, which numbers have trodden, we may find a way at length into her inner chambers.”

“For you have but to follow and as it were hound nature in her wanderings, and you will be able, when you like, to lead and drive her afterwards to the same place again.”

[Science and technology do not] “merely exert a gentle guidance over nature’s course; they have the power to conquer and subdue her, to shake her to her foundations.”

But, while Nature is a shy female waiting to be seduced and conquered, we also recognize that Nature is a powerful, almost God-like force. Tellingly, when Bacon talks about this aspect, the metaphorical gender switches, and now Nature is all too male:

“as if the divine nature enjoyed the kindly innocence in such hide-and-seek, hiding only in order to be found, and with characteristic indulgence desired the human mind to join Him in this sport.”

So much meaning lurking in a few innocent pronouns! We like to pretend that the way we do science, and the way we conceptualize our activity, is more or less inevitable; but there are a lot of explicit choices along the way.

  • Aaron Bergman

    The third person/first person has a reasonable rule behind it (which I confess to not always following). “We” refers to the author and the reader while “I” refers solely to the author. In other words, one might write “we can see that” and “I will show that” in the same paper.

  • http://blogs.discovermagazine.com/cosmicvariance/sean/ Sean

    That’s true, but different; there are a large number of single-author papers (the vast majority, if my impression is correct) in which the word “I” is always replaced by “we,” whether appropriate or not.

  • http://www.sunclipse.org Blake Stacey

    When we had to write term papers in third-semester QM, Prof. Rajagopal addressed this specifically:

    Feel free to use whichever voice you are most comfortable with. “I will show,” “we will show” or “it will be shown” are all fine. For unknown reasons, some students seem to think that personal pronouns are banned and the passive voice is required. Nothing could be further from the truth. Good scientific writing should be animated and compelling. Your paper should “tell a physics story”. I find the overuse of the passive voice to be deadening. Don’t be dull. Clarity and precision come first, but don’t fall into the trap of thinking that this can only be accomplished via boring your reader to tears. Not true.

  • http://www.sunclipse.org Blake Stacey

    I wonder who will be the first prescriptivist in this thread to castigate Aaron Bergman for splitting an infinitive and Rajagopal (in absentia) for ending a sentence with a preposition. Now that’s the sort of nonsense up with which we should not put. . . .

  • http://www.mpe.mpg.de/~erwin/ Peter Erwin

    Aaron has a good point about the “pedagogical” we — the sort of “we” a lecturer might use in stepping through an argument (“So we can see here that…”). But I agree with Sean that there are a lot of single-author papers where “I” is never used, even when it would be the logical choice (“I used the following data reduction technique…”).

    When I’m reading a paper like that, there’s a mischeivous part of my mind that wants to know if the author thinks he or she is royalty…

  • http://www.leekottner.com Lee Kottner

    Passive voice is a common problem in beginning writers, but most writing teachers are more interested in excising it than in nurturing it. Scientific prose is the only exception and it’s always seemed absurd to me, objectivity (which is a lovely myth anyway) or not.

    “What would Strunk and White say?” AAAIIIIEEEEE! That’s what.

  • Rien

    Agree completely. When I wrote my first single-author paper I started the abstract with the word “I”. I hope I also did split some infinitives and I’m sure I ended some sentences with prepositions. I also tend to wonder if authors are using the royal we.

    Here’s a good take on writing papers by the way:

  • http://quasar9.blogspot.com/ Quasar9

    “I invite all such to join themselves, as true sons of knowledge, with me, that passing by the outer courts of nature, which numbers have trodden, we may find a way at length into her inner chamber.”

    lol Blake, I guess we should hope no one asks how can knowledge have sons.
    I hope that no feminists object, and are appeased by Nature being feminine

    Though “finding a way at length into her inner chamber” – had me in stitches
    could it also be allegorical – as in of things (or strings) to come!

  • http://www.mpe.mpg.de/~erwin/ Peter Erwin

    On the subject of Bacon — I’m not convinced it’s appropriate to lay all the blame on him. For one thing, he was writing at a time when most scientific reports were still done in Latin, so it’s not clear how style in Latin would necessarily carry over to style in, say, English. (And did papers in his time really have “abstracts” as we think of them?)

    For another, it’s my impression that first-person-singular scientific writing continued for quite some time after that. For example, Newton’s paper on splitting light with a prism is full of the first person singular, and even the occasional personal digression (“Amidst these thoughts I was forced from Cambridge by the Intervening Plague, and it was more then [sic] two years, before I proceeded further.”) A quick look at some online scientific works by Charles Darwin and James Clerk Maxwell suggests both of them were perfectly happy using “I” as well, two hundred years after Newton.

    In fact, Darwin and Maxwell appear to use both first-person-singular and the pedagogic “we”, as in this passage from Darwin’s book on coral reefs: “… as we have just seen, the presence in most cases of upraised organic remains of a modern date. I may here remark that the reefs were all coloured before the volcanoes were added to the map, or indeed before I knew of the existence of several of them.” (From Darwin’s book on coral reefs.)

    I suspect the pervasive emphasis on passive voice and first-person-plural may be more of a 20th Century phenomenon.

  • http://tyrannogenius.blogspot.com Neil B.

    The editorial community does not all think alike, and there is some common sense out there:
    In the AIP Style manual, IIRC, they actually say not to indulge in “we” nor in passive, I think, saying it is pointless, stilted, etc.

    BTW, I need to know: which publishers/Editors expect the forced passive/plural usage, and which do not?

  • jepe

    great post!
    The obsession with passive voice in scientific writing is frustrating after being punished repeatedly in Writing courses for daring to use the passive voice. It’s a schism I bring up repeatedly w/my Humanities colleagues.

    Insistence on the passive voice can make scientific writing quite distasteful. Nevertheless, there seems to be this feeling one must write passively otherwise the content will not be taken seriously. There is a sense among students that being taken seriously demands phrases like: “It is easily seen that…A sharp increase is observed…” Stereotypical images of the prof w/the tweed jacket, or the lab rat w/the white lab coat come to mind.

    The compromise for our stuff is: Materials and Methods are written in the passive voice. Write the meat in the active voice.

  • Sophia
  • http://www.astro.caltech.edu/~aam ashish

    We treat the “We” used in single author papers as the royal version of I and we find it preferable in general.

  • http://thecrossedpond.com adam

    I had to read Bacon’s Novum Organum once (although not in latin, thank Christ). It wasn’t the highpoint of my life.

  • http://www.pieterkok.com/index.html PK

    I wrote a single-author paper for IEEE in which I used both “I” and “we” in the sense described above. The editor changed all the Is and some random wes into “the author”, as well as turning many active sentences into passive ones. Hilarity ensued, with gems such as: “The author needs single-photon detectors”.

  • Aaron Bergman

    Not that there’s anything wrong with splitting infinitives — I’m all for it — but I don’t think I split one. “Following” is a gerund serving as a prepositional object to the verb “confess”.

  • http://countiblis.blogspot.com Count Iblis

    I also used “we” instead of “I” in my last single authored paper. I did think about this while writing the paper, but I found that replacing “we” by “I” everywhere made the article read awkward.

  • jepe

    Sophia– Thx for the link. I should’ve known. I was lucky to have had Mermin for classical mech, and it was a fantastic experience. His book, “Boojums all the way”, has good commentary on how to write…and how not to write.

  • Cynthia

    Unless you’re a woman who truly enjoys being conquered by a man, I just can’t imagine any woman (in her right mind, that is) wanting to sleep with a man with Francis Bacon’s mindset!

  • http://blogs.discovermagazine.com/cosmicvariance/julianne Julianne

    I wrote my most recent single author paper in the first person singular, and was told by the referee to change it because I sounded egotistical.

    We were not amused.

    But text was changed.

  • Dave

    We have multiple personality disorder, so first person plural is always appropriate.

  • miller

    If professional scientists (such as yourself) agree that it is pointless to avoid first person in science writing, would you please inform high school science teachers. It was apparent to everyone in my class, as well as the teacher himself, that it was a pretty stupid system. But I was under the impression that that’s just how it’s done, so we had to go with it. Luckily, I never got good at it.

  • http://thecrossedpond.com adam

    Cynthia #19: It was widely said at the time, I believe, that Bacon would himself have preferred not to lie with woman. Everybody wins!

  • http://www.sunclipse.org Blake Stacey

    Aaron Bergman (#16):

    You’re right.

    One day, I’ll learn to think before typing. Until that day comes. . . .

  • http://www.bioephemera.com bioephemera

    The passive disease afflicts grant offices, too. I wrote a classroom technology grant for $50K – quite a sizeable one for our small college. Since it would be reviewed by politicians who didn’t have science backgrounds (or a lot of patience), I wrote it in jargon-free, concise first person. Our college grants admin had this feedback:

    Admin: “It sounds completely unprofessional. Write it in third person – that’s how all professional documents should be written.”

    Me: “Changing the grant to third person will require passive voice, and introduce unnecessary wordiness in several places. Given the audience, I think it’s a bad idea.”

    Admin: “Ha ha. No, it won’t. If you were an English major instead of a scientist, you’d know that third person and passive voice NEVER introduce wordiness to a document.”

    Me: “I have a BA in English.”

    Admin (whose degree is in Education Administration): “Well, if I was you, I wouldn’t admit to it.”

  • Jason

    The third person is usually simpler and shorter than the first person. For instance, when stating a definition, it is shorter to write “A tensor is SYMMETRIC if …” than to write “We will say that a tensor is SYMMETRIC if …” I was taught this is the reason to prefer the third person.

  • Illirikim

    I really don’t care about all these rules, but I *would* like to see more attention paid to elegance and variety. I saw a paper recently in which the author writes simply: “It would have been reasonable indeed to suspect that [something] holds in all cases. One sees now that it is not so.” Nice!

  • Cynthia

    adam: well, then, let Bacon do his thing! ;-)

  • http://jamesdmiller.blogspot.com/ James D. Miller

    I’m an economics professor and recently had a student turn in a paper in the passive voice. I asked her why she did this and she told me that her Engineering professor said this was how papers are supposed to be written.

    If any of you science people are telling students to use the passive voice in their writings then do your students a favor and let them know that non-science professors hate this.

  • Clearly then bjormn!

    While this post reminds me of an article I read a few months ago, which perhaps a nudge the topic of the post onto a nearby branch, I feel it is important to bring it up.

    Apart from the entire speech/article being brilliant (and brief!), this one particular passage I think is especially interesting in the context of this discussion:

    When I give a lecture I want my students to know that something is happening in real time—that I am thinking through the argument as it is delivered, and responding continually to feedback, verbal and otherwise, from the audience. Personally, I never bring notes to a lecture unless I am egregiously ill-prepared, for they break a very delicate and important bond of trust with the listener: If B really follows from A, how come he has to refer to his notes?

    Millikan Lecture 1997: Is there a text in this class? David J. Griffiths, Am. J. Phys. 65, 1141 (1997).

    I’m sure all of us have experienced first hand this exact sort of thing David touched on. I find it is surprisingly frequent that a lecturer, or perhaps a text even, will say something like “clearly [insert extremely esoteric and challenging concept here] follows from [insert another extremely esoteric and challenging concept here], and as a result it is clear that flying spaghetti monsters DO exist!”

    The problem I have with this usage is that the word “clearly” implies to the audience that they should be able to grasp this implication very easily, where instead what may have been meant was something more along the lines of “Although perhaps exceedingly difficult to grasp, once you do fully understand the topic at hand it may then become clear to you that ‘[insert extremely esoteric and challenging concept here] follows from [insert another extremely esoteric and challenging concept here], and as a result it is clear that flying spaghetti monsters DO exist!’”

    So then, can anyone thing of an appropriate substitute for the word ‘clearly’? Maybe I’ll go ask David tomorrow or send him an email or something. I feel that this branch I bring up relevant to the discussion because just by using the word clearly in this way, it can be quite discouraging to otherwise perfectly good students who then question their own aptitude because they may think “Wow, how am I supposed to be a physicist like this guy if I can’t even do the things he thinks are really easy!” where in fact pretty much no one finds it easy.

    More on point though, where do you think presentations or speeches fall in the great first person, third person divide, or are speeches, presentations, lectures etc. considered more informal and as a result both are just fine? This discussion also reminds me of the formalism Hilbert was apparently famous for.

    I apologize for such a lengthy comment.

  • cynic

    George Orwell’s five rules for effective writing include the eschewal of the passive voice, and are appropriate to published scientific discourse. Especially the sixth rule of course.

  • Pingback: First-person pronouns in science writing « Entertaining Research

  • Rien

    Orwell’s rule 3 is important too: “If it is possible to cut a word out, always cut it out”. Thus, “clearly” can clearly be removed from all scientific writing.

  • http://www.bioephemera.com bioephemera

    Jason: in your tensor example, I think the two sentences suggest two different meanings. Stating the definition outright (“a tensor is. . .”) implies it’s an established definition. But the latter (“we will say that a tensor is. . .”) implies to me that the definition is new, because you are calling attention to your action in creating it. If you don’t mean to imply that the definition is new, then I agree the latter is unnecessarily wordy.

    Anyway, it depends on your purpose and audience, but in my experience, statements like “I will include these computer activities in my labs” and “Our students do not have access to modern equipment” are more direct and reader-friendly than “Participating faculty will include these computer activities in their labs” or “Students in the affected courses at University X do not have access to modern equipment.” In a part of the country where “intellectualese” is distrusted, I try to be plain-spoken, especially when asking for money. :)

  • http://www.amara.com/ Amara

    About Bacon’s personification of Nature:
    It doesn’t get any better than D.H. Lawrence’s description of Mt. Etna in _Sea and Sardinia_ (1921):

    “With her strange winds prowling round her like Circe’s panthers, some black, some white. With her strange, remote communications and her terrible dynamic exhalations. She makes men mad. Such terrible vibrations of wicket and beautiful electricity she throws about her, like a deadly net! Nay, sometimes, verily, one can feel a new current of her demon magnetism seize one’s living tissue and change the peaceful life of one’s active cells. She makes a storm in the living plasm and a new adjustment. And sometimes it is like a madness.

    This timeless Grecian Etna, in her lower-heaven loveliness, so lovely, so lovely, what a torturer! Not many men can really stand her, without losing their souls. She is like Circe. Unless a man is very strong, she takes his soul away from him and leaves him not a beast, but an elemental creature, intelligent and soulless. Intelligent, almost inspired, and soulless, like the Etna Sicilians. Intelligent daimons and humanly, according to us, the most stupid people on earth. Ach, horror! How many men, how many races, has Etna put to flight! It was she who broke the quick of the Greek soul. And after the Greeks, she gave the Romans, the Normans, the Arabs, the Spaniards, the French, the Italians, even the English, she gave them all their inspired hour and broke their souls.”

  • Thomas Larsson

    The answer is known to these bloggers.

  • Mark Hudson

    Isn’t there a “myth” (I have no idea if it is true or not, but I tend to assume that myths are not true) along the lines of [famous particle physicist] being told that he had to use “we” even though he was a paper’s sole author. So he added his cat as co-author?

    I may have got that slightly garbled…

  • http://www.mpe.mpg.de/~erwin/ Peter Erwin

    Jason @ 26:
    The third person is usually simpler and shorter than the first person. For instance, when stating a definition, it is shorter to write “A tensor is SYMMETRIC if …” than to write “We will say that a tensor is SYMMETRIC if …” I was taught this is the reason to prefer the third person.

    Of course that’s shorter. And that’s a perfectly sensible way to write; no one (I think) is suggesting “Third person bad, first person good!”, or that all definitions, simple statements of fact, etc. must use the first person!

    The problem arises when people (or the editors they deal with) think they need to elminate all trace of the first person from their writing, even when it’s the more natural and appropriate choice.

  • http://www.mpe.mpg.de/~erwin/ Peter Erwin

    Before we go overboard on dissing the passive voice, and on praising Orwell’s writing advice, it’s worth taking a look at this post on the subject from Geoff Pullum, summarizing some discussion among the linguists at Language Log. Pullum notes a “delicious passage” in Merriam-Webster’s Dictionary of English Usage on the subject:

    Bryant 1962 reports three statistical studies of passive versus active sentences in various periodicals; the highest incidence of passive constructions was 13 percent. Orwell runs to a little over 20 percent in “Politics and the English Language.”

    So Orwell seems to honor his own law more in the breach than in the observance.

    In a later post, Mark Liberman remarks that a quick analysis of E.B. White’s writing indicates a passive-voice frequency of about 20%, and shows how Winston Churchill (“a model of forceful eloquence”) used an even higher frequency in a dramatic narrative about 19th Century colonial wars.

    The passive voice isn’t in itself good or bad, even if it does get overused sometimes in scientific writing.

  • Pingback: Scientific writing: Too impersonal? | Compactified Realisations

  • Jason

    Bioephemera — In my writing I rarely need to distinguish between an established definition and a new definition. If I do need to point out a definition is established (for instance because I am redefining it), I do so with a citation.

    Peter — Of course I believe in using the first person when it is appropriate. For instance it is appropriate in an introduction or conclusion when explaining motivation, history of the problem, etc. But at least in my field, in the body of the article (i.e., when presenting new results), it is almost never appropriate. Authors often believe it is appropriate — I was one of these for many years. But then, at the urging of an older colleague, I played the game of removing all sentences with “I” or “we” from the body of one of my preprints. It didn’t change the true content of the article, and it made the article significantly shorter.

  • http://thecrossedpond.com adam

    Personally, I very much dislike the use of ‘I’ in papers. I do appear to be in a minority, however, at least here.

  • http://www.pieterkok.com/index.html PK

    why, adam?

  • http://www.mpe.mpg.de/~erwin/ Peter Erwin

    Jason @ 41: I suspect we’re talking about slightly different things. If you could actually remove the I-containing sentences you refer to without damaging the content or the clarity, then they were arguably superfluous and deserved to be removed. (Maybe the resulting text is shorter but less entertaining now; maybe it’s more concise and readable.) The fact that they used the first person is irrelevant, unless there was something about your particular writing style that made irrelevant sentences more likely to be first-person…

    The issue that Chad and Sean are referring to is not removing first-person sentences — at least not if they contain important information — but the idea that they must be rewritten so that “I” (or “we” even if it’s a multi-author paper) never appears. For example, this set of instructions on preparing abstracts for meetings of the Acoustical Society of America contains the following advice:

    Use passives instead of pronouns “I” and “we,” e.g., “It was noted” instead of “We noted.”

    which in this case makes things slightly wordier, the opposite of what you want for an abstract…

  • http://thecrossedpond.com adam

    PK #43: It’s just a personal thing, I guess. I don’t mind the pedagogical ‘we’ and I imagine that my preferences probably have no logical basis.

  • http://arunsmusings.blogspot.com Arun

    “Mistakes were made.”

  • http://countiblis.blogspot.com Count Iblis

    A referee of a single authored paper I wrote some time ago actually consistently referred to me as “the authors” in the report :)

  • Haelfix

    In my experience, if a referree returns a paper admonishing this sort of thing, it usually means he hates the paper but can’t find sufficient reasons to reject it. In other words, hes just being snooty. In practise this just means conforming so that they aren’t given that chance.

    I’ve never understood why the ‘royal we’ (which is what it sounds like most of the time) has to be used in science writing, it strikes me as archaic and a tradition that confuses things rather than elucidates them.

    In fact, i’d love to see more papers where there is explicit disagreement between co authors. (eg Witten things that this 2 form has to be closed, but Vafa disagrees). ‘We’ tend to learn more from disagreement in Science than pure consensus, even if its regarding unimportant technical details.

  • http://thecrossedpond.com adam

    I don’t think that most referee are looking for a reason to refuse a piece, Haelfix. It seems to me to be more likely that they either have an opinion as to the correct style, or possibly are looking at journal guidelines (if such are available).

  • Michael D

    And to Peter’s posts from Language Log, here’s another example of where the demand for use of active voice, doesn’t work out all that well.



  • http://www.pieterkok.com/index.html PK

    I think it is perfectly fine to use the passive voice when the acting agent is irrelevant to the point you try to make. For example, it is much better to say “The samples were cleaned with chloroform” than “Joe Blogs cleaned the samples with chloroform”. Who cares whether it was Blogs or Smith?

    I also think that the passive voice often acts as glue that keeps a piece of writing together. Instead of all active sentences each demanding the attention of the reader, passive sentences are resting points. But of course you do not want a text completely composed of resting points.

  • Pingback: æ¡‘æž—å¿— » 科技写作的一些潜规则


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Cosmic Variance

Random samplings from a universe of ideas.

About Sean Carroll

Sean Carroll is a Senior Research Associate in the Department of Physics at the California Institute of Technology. His research interests include theoretical aspects of cosmology, field theory, and gravitation. His most recent book is The Particle at the End of the Universe, about the Large Hadron Collider and the search for the Higgs boson. Here are some of his favorite blog posts, home page, and email: carroll [at] cosmicvariance.com .


See More

Collapse bottom bar

Login to your Account

E-mail address:
Remember me
Forgot your password?
No problem. Click here to have it e-mailed to you.

Not Registered Yet?

Register now for FREE. Registration only takes a few minutes to complete. Register now »