Girl, Interrupting

November 10, 2010

It’s not about you, its about the data!

Or why I think the passive voice is actually GOOD for science

There are advocates of ‘plain language for scientists’ for example Harvard Health blogs who suggest Med journals should write more like Micheal Crichton and Evidence Soup who wants medical journals to ‘stop with the passive voice, already’. The summary being that scientists write far too much in the passive vs. the active voice (in English that is) and science would be easier to understand if it were written in the active voice. There are also suggestions that ‘passive-voice science writing’ is elitist.

I disagree on both counts but first a quick grammar review, what is the difference between active and passive?

In the active voice the object receives the action of the verb as in:

“Cats eat fish (active).”

In the passive voice the subject receives the action of the verb:

“Fish are eaten by cats (passive).”

Which is great for cats and fish but what about scientists?

One of the arguments for ‘active English’ in science writing is that if English isn’t your first language, reading in the passive voice is more difficult. A couple of French, Bangladeshi and Dutch friends have confirmed this. On the other hand, an Italian scientist friend of mine told me it is actually easier to read in the formal passive voice, because it is closer to the way science is written in Italian. In her opinion, ‘plain English’ initiatives never help non-native English speakers, she believes it is mostly for those of us that already speak English. So I would say just from my unofficial straw poll that active English being easier for foreigners is still up for debate.

Regardless of which is easier to read, the passive voice is a construct of English, not of science and for better or worse English has become the de facto lingua-Franca of modern science, at least in the West. If we based the ‘language of science’ on the most abundant language in world-wide science, that would be Chinese, which may be preferable as Chinese has no verb tenses.

And English, well, is odd. English is written and indeed spoken in a complex combination of active and passive voices. This doesn’t only happen in science articles; this even happens in the Daily Mail. Take for instance the following excerpt from an article published in the Daily Mail:

EastEnders star Steve McFadden has been arrested and bailed (passive) over claims that he harassed a woman (active) believed to be the mother of his baby girl.

The 51-year-old actor was picked up by police in Haringey, north London, on Wednesday (passive).

After being questioned by officers he was released on bail to return to a police station early next year (passive) .

It is understood that McFadden – who plays EastEnders hard man Phil Mitchell – was arrested following a complaint by former partner Dr Rachel Sidwell (passive).

The pair have a daughter, Amelie Tinkerbell, now 17 months old

Now if you translate that all into the active this is what you get::

Police have arrested Steve McFadden. The court released McFadden on bail. The police and the court acted on claims that McFadden had harassed a woman believed to be the mother of his baby girl.

The police picked up the 51 year old actor in Haringey, north London, on Wednesday.

Police questioned Steve McFadden. The court released McFadden on bail to return to the police station early next year.

It is understood that the police arrested McFadden – who plays EastEnders hard man Phil Mitchell – following a complaint
by former partner Dr Rachel Sidwell.

The pair have a daughter, Amelie Tinkerbell, now 17 months old.

OK so this is ‘plain english’ but in the active version the story becomes all about the police not about Steve McFadden which is what the story is intended to be about – Steve McFadden is the object receiving the action by the police, who are the subject.

And this is the point! As scientists we teach our students to write about the data – which are the object of any experiment. Why? Because the science is about the data and the data are INDEPENDENT of who did the experiment (well except if there is fraud). Science is about the physical world around us which is exhibited by the data and not the person doing the experiment.

Explicitly, in technical science journals you will see things like:

‘The data were collected’
(passive) which is about the data.

Rather than ‘I collected the data’ (active) which is about YOU collecting the data in the first person
or if you prefer the 3rd person ‘The post doctoral researcher collected the data’ (still active) which, again, is about the person collecting the data.

And about the charge of being ‘elitist’ I would say no more than the English language is elitist. Let me repeat, data should be written about as if it is independent of the people doing the experiment. Science is and should be about the data and it is damn nigh impossible to write about data in the active voice because data don’t collect themselves. (N.B. Data is plural, unlike Data the Star Trek dude who is singular) Of course someone collects it and makes the figures for papers, but, again, science isn’t about the researchers, its about the data.

If you are a native English speaker I challenge you to speak only active English for the day. Or even write entirely in the active voice. In reality science papers, like most English writing is a hodge-podge of active and passive English.

Maybe advocates of the active voice in science writing really need to just teach scientists better English writing skills. I would argue that perhaps technical scientific papers are difficult to read because they are simply badly written regardless of voice.



  1. […] This post was mentioned on Twitter by Elmira Bayrasli, Sylvia McLain. Sylvia McLain said: Its about the data not about you! or why I think we should DEFEND passive voice in science journals my new blog post […]

    Pingback by Tweets that mention It’s not about you, its about the data! « Girl, Interrupting -- — November 10, 2010 @ 2:49 pm | Reply

  2. @quote: In her opinion, ‘plain English’ initiatives never help non-native English speakers

    I agree. The ‘plain English’ guidelines that I know about do not say much about how to write for non-native English speakers. (For guidelines about how to write for non-native English speakers, I recommend Kohl’s ‘Global English style guide’,

    @quote: science would be easier to understand if it were written in the active voice
    Nonsense. Science is difficult. Writing scientific texts in the active voice does not make science easier.

    @quote: There are also suggestions that ‘passive-voice science writing’ is elitist.
    The word ‘elitist’ is vague. What does ‘elitist’ mean in the context of science writing? Elitism can be good or bad, depending on a person’s perspective. (Think about ‘best in class’ compared to ‘snobbery’.)

    @quote: Maybe advocates of the active voice in science writing really need to just teach scientists better English writing skills. I would argue that perhaps technical scientific papers are difficult to read because they are simply badly written regardless of voice.

    I agree. Frequently, I read scientific texts and technical texts that are not clear. For example, “…after which the dried solid is at least virtually completely separated from the zeolite granules…” Possibly, the writer means, “after which almost all the dried solid is separated from the zeolite granules”. Possibly, the writer means something different.

    Comment by Mike Unwalla — November 11, 2010 @ 8:49 am | Reply

  3. I was directed here by a mutual friend.

    This blog post contains the majority of my complaints about not only requests made to “dumb-down” scientific studies, but also a few of my grammar pet-peeves.

    Mind if I link to you on my blog occasionally?

    Comment by Squeaky Wheel — November 16, 2010 @ 7:38 pm | Reply

    • No of course not, I would be flattered, link away! And thanks

      Comment by sylviamclain — November 16, 2010 @ 7:46 pm | Reply

  4. Long time lurker, first time commenter here. 🙂

    I have to go out on a limb and disagree here, in part. First, where I agree: I’ll completely agree that “plain English” initiatives can be very biased toward non-English speakers. A related problem (which you hint at) is the dominance of English as the language for scientific study. You see this, for example, in Spanish-language textbooks–which are often out-of-date due to the delays introduced in the translation and republication process. The effect is to make it difficult for non-English speakers to gain a foothold in many scientific disciplines. Plain English does become important, for example, in social science and medicine where the subjects of the research arguably have a right to read what is being written in terms that they can understand. I also completely agree that the passive voice is overly maligned.

    Where I disagree is in the idea that data are independent of who is conducting the experiment. I would submit that this is rarely ever true. I’m not arguing here that objectivity is a bad goal. But, researchers inevitably make hundreds of personal choices, many of which are not documented in their methodologies because they are not considered significant. Also, the data never speak for themselves–scientists always do the speaking. (In practice, this works quite well most of the time.) So, in this case, I have to still favor the active voice: it isn’t enough to say “data were collected”–I want to know who did the collecting, and sometimes even whose payroll he or she was on. I guess at the root, I disagree that science is about the data; scientists collect data in service of larger paradigms, which do not change based on the results of experimental testing. They only change when one generation of scientists die and are replaced by their successors, or when there is a major social upheaval. I think a failure to grasp this on the part of some scientists–thinking of climatologists here–is one reason for the recent swell of anti-science sentiment in the public sphere.

    Comment by Ted Maclin — November 16, 2010 @ 9:25 pm | Reply

    • Hi Ted thank you for your comment.
      Perhaps I can clarify:

      The physical world is indeed independent from you as a human. For instance if you, I and your grandmother go to the top of the tower of Pisa and drop an apple off that tower, it will fall to the ground. At least no one has ever observed(disregarding wind) the apple say going up into the atmosphere and then exploding. This is INDEPENDENT of who dropped the apple.

      Following on with this example, if me and granny drop the apple straight off and then you hoik it up in the air first, our apples will hit the ground first and yours last. Meaning that me and granny are collecting similar data and you are collected different data. Now i would agree with you, that people don’t always do the same thing when collecting data, even though they may think that they are. But in a ideal world we can speak to each other about that – we follow certain protocols and for the most part if we do the same thing, we get the same answer (within error) – if you can’t repeat an experiment you are either not following the protocol correctly or it isn’t presented to you correctly.
      This is a constant problem, literature not really being clear about what was done and i would agree that scientists need to get better at this, but it has nothing to do with the passive voice. Also there is the point of brevity, if you work in a particular field you tend to use jargon and references to how you did what, otherwise all papers would be about 100 pages long which would make clarity even worse.

      What you are referring to is a separate issue. You are referring to the performance of the experiment by the researcher (such as in the apple example above) and the misinterpretation of that data, these are separate issues and indeed separate issues from each other. If you collected data under x-conditions, these data should be reproducible. If it is not, then there is a problem (see above) or you have analyzed your data improperly.

      Paradigms are not data, they are interpretation of data. Data is data. You can have false data, bad data, erroneous data, collected your data inappropriately, etc. but this is not the same as the interpretation of this data or indeed the method by which you obtain the data.

      Paradigm shifts happen slowly – but they do happen. I would agree that scientists do get stuck in paradigms (unconscious bias) but this still is not the same thing as the data not being the data. Sometimes people over-interpret their data – or make claims based on data that they shouldn’t be making, but scientists are after all human and yes they do collect data making certain assumptions .. but not always – There is a great book about this by the way called the Mismeasure of Man by Stephen J. Gould you should read it. It is also true that scientists will ignore data that doesn’t fit their model or pet theory. But this is about science being practiced badly not science itself.

      Science is a human construct by which we interpret the natural world – whether this discipline is physics, chemistry or physchology. Its accuracy and effectiveness depends on alot of different factors and what questions you are asking.

      What i think is bad about the way science is presented, is that it is presented as an absolute truth very often, which is what I think you are referring to. This is bad science. Even the laws of physics are not really laws in the same sense as people think of laws. they are not immutable, look at Newton’s laws, which were overturned by quantum mechanics, but still describe quite well the macroscopic physical world.

      In a technical paper, you do know all of the things you want to know… They have authors, the authors declare their funding. The results are presented and then interpreted.

      Being stuck in your own paradigms and not listening to new data which causes an anti-science sentiment is fair. But people should be pissed off because scientists aren’t doing their jobs properly, not hate science itself.

      I also think it should be noted that some of the worst scientific gaffs (remember cold fusion?) were from people who spent alot of time presenting in the first person – We did this, we did that, we discovered cold fusion…. Using the active voice in no way helps with this problem…

      Thanks again for your comment!

      Comment by sylviamclain — November 18, 2010 @ 6:13 pm | Reply

  5. Hi, I’m kind of going to agree with Ted here. I recommend reading Montgomery’s The Scientific Voice, because he explains it much better than I can. But the passive voice disguises the agency of scientists – it makes it sound like the word of god – this was done, these results were seen.

    In a sense it’s a rhetorical trick, a sleight of hand, which makes us forget that human beings, with all their fallibility, biases and preconceptions, did those experiments, observed the results, even thought of the question to investigate in the first place. Science strives to be objective, but there’s always some subjectivity. Surely the best scientists bear that in mind and try to account for it or control for it, or keep an open mind about their conclusions?
    I think the passive voice helps people to forget that, whereas the active voice reminds them.

    BTW, I agree that ‘plain english’ is not always best for non native speakers. Years ago I did TEFL training and it’s amazing how much what we’d think of as simple (and would use, eg speaking to a child) is actually very grammatically complex, or idiomatic. Formal language can be much clearer.

    Comment by Sophia Collins — November 18, 2010 @ 6:41 pm | Reply

    • I kinda see this point, but I disagree, I think active voice makes it MORE about the scientist than the science…

      But what I do think is that the unifying thing here between all of our opinions, is that we all agree that science should not be presented as if it is the ultimate answer, the end all and be.. As it often is I think, especially in the news… And some scientists in the media definitely do this, it personally bothers me a lot!

      I am a scientist, and I write in both passive and active voice, and I do try to be as objective as I can, whether I do or not is another question… I am aware I have biases ( see my post from Tuesday on unconscious bias )

      I would also like to add that I am talking in this post purely about technical research papers, NOT media articles, press releases, seminars….

      Thanks for commenting Sophia

      Comment by sylviamclain — November 18, 2010 @ 7:16 pm | Reply

  6. Great discussion; I think we are in agreement in large part. I may write more about this on my own blog. I think as I write more that you are talking more about the conduct of science and I am talking more about the general practice of science.

    Where I think we differ is that I don’t see the physical world as independent of me as a human. Now, that is a provocative statement, so let me clarify. I’ll approach this from two directions.

    First, the idea that the world is not separate from us. I’m not denying physical reality–but I come from a critical realist approach. In the apple-off-the-tower example you cite above, the apple wouldn’t ever fall off the tower if you, or me, or your grandmother didn’t take it there and drop it. Humans conduct experiments, and more importantly humans decide what counts as data, good data, bad data, accurate data, and so on. We base those choices on culture, history, technology, and other human factors. Without humans, data would not exist. Sure, apples might still grow on trees (though possibly not, since humans have played such a huge role in the evolution of apples), but they wouldn’t be called apples and nobody would be measuring how fast they fall. Ok, aliens or other intelligent species might, but you get the point. This may sound like I’m just playing semantics, but I think it is important: I’d argue that all data is a human contrivance to some degree. When we write in the detached passive voice, the data gains an implied objectivity that may or may not be justified. The data seems to exist on its own, and while apples may do that, data about apples doesn’t. I look at one of my favorite scientific works, Darwin’s Origin of Species–and it is largely written in first person. The idea that science shouldn’t be done that way is a 20th-century development. Others like Kuhn and Polyani have written a lot about the ways that culture, society, and personality influence the development of science, that all scientific knowledge is at least partially subjective. There is a physical world out there, but the world as we understand it is always a human construct.

    Second, the idea that we are not separate from the world. Sociologists, philosophers, biologists, and computer scientists have been coming to surprising consensus on this point. The physical, mental, and social construction of the brain doesn’t just influence what we believe: it influences what we can think. I like some of the recent writing on the idea of the extended mind. The idea is that “mind” (versus brain) is not bound by the confines of our skulls. The famous hypothetical case is of a man with limited short term memory who uses a notebook to keep up in daily life; in that case, where the notebook serves the function of mind, the notebook counts as part of the mans mind. Take that idea and expand it outward, and papers, tools, computers, particle accelerators, other people, religious groups, and so on all count as part of the mind because we use them as part of our cognitive processes. Not in a metaphorical way–since we use external things as part of our cognition, they are not neutral in the operation of our cognition. Real-world objects like trees, rocks, neutrons, and stars, also act as reservoirs of cultural symbols. The bottom line here is that the physical world and our human relations to it influences what questions we ask, how we go about finding answers, what we count as “data,” and even how our identities and relationships develop.

    Many scientists respond to these sorts of statements by saying that science takes everything into account through experimental controls, attention to methodology, ensuring repeatability. But not all science is based on experimentation and I’d argue that experimentation is not the most important part of the project, since the experimental design depends on the nature of the hypothesis developed by the scientist. The nature of that hypothesis is always mediated by cultural and social factors.

    All this talk about apples has me craving pie. Now, that is a cultural construct I can get into!

    Comment by Ted Maclin — November 18, 2010 @ 7:47 pm | Reply

    • I will answer this in kind soon – I will say quickly I do think we largely agree about some of these things. and of course i am part of the physical world but then we might be able to get into a discussion about uh physics vs. so-called softer ‘sciences’ (not a word I like) – but I think you are opening up lots of (very interesting) issues which are not really what my post is talking about – which means perhaps I have been unlcear… will try to clear it up soon..

      but on a flippant note – have you ever read the Origin of Species all the way through? I bet you have, but that is one damn turgid book – I think….

      Comment by sylviamclain — November 18, 2010 @ 7:57 pm | Reply

    • Hi
      Can I get a quick clarification before I attempt to reply?
      You say ‘would the apple fall if we didn’t drop it’. 
      Of course not, but part of science is conducting experiments which mimic the physical world to understand it’s properties…

      You seem to me to be saying almost experiment is not a part of science…

      Experimental set ups can be crap in that they don’t mimic the real world or are designed to fulfil a preconceived notion.. To prove a ‘pet theory’ instead of being a true hypothesis test..

      Karl Popper defines something as truly science if and only if it has the potential to be falsifiable, which I think has a lot of truth in it..


      Comment by sylviamclain — November 18, 2010 @ 8:33 pm | Reply

  7. I wouldn’t go so far as to say that experimentation isn’t part of science, though some sciences use it more heavily than others. Experimentation is in my mind secondary to hypothesis formation as the key scientific process, since without hypothesis there is no experiment. But, humanity comes into play even in the experiment. You say that ‘Experimental set ups can be crap in that they don’t mimic the real world or are designed to fulfil a preconceived notion,’ but dropping an apple (maybe not the best example but I’ll run with it) as an experimental design says something about preconceived notions of how the world works. An scientist who has never seen an apple tree may drop a snow ball, or a banana–which should yield nearly the same results if they were working in a vacuum, but they aren’t. Falsification is all well and good as long as we realize that a single hypothesis can never be tested because it (the hypothesis) is always wrapped in a warm blanket of theories, worldviews, and cultural norms…

    Comment by Ted Maclin — November 18, 2010 @ 9:50 pm | Reply

  8. I don’t really agree with that, yes hypothses are culturally biased to some extent, some more than others

    It doesn’t matter what you drop though gravity does exist as do many phenomena we do ‘name’ the names our ours but the phenomena still exists..


    Comment by sylviamclain — November 18, 2010 @ 11:33 pm | Reply

  9. I’m going to totally sidestep the metaphysics flying around and go straight for the language: the passive voice is ugly and ungainly and while I wouldn’t say “it should never be used”, I do think its use should be reduced. That is to say, we should use it less. I’m willing to class that as a subjective statement…

    The implied subtext in using one or the other is, IMHO, pretty marginal. Either you say “we dropped an apple (now why don’t you try it at home and tell me whether you find the same thing?)” or “an apple was dropped (a universal apple, by me, with my set of methodological quirks)”. If you’re aware of the inherent conflicts between consensus and individual experience, you will understand the parenthetical bit of each if the above statements; and if not, you take away that some person designed an experiment where an apple was dropped. Changing the language will not illuminate much about individual bias or collective consensus, certainly less than reading a book. And it will not illuminate a researcher’s methodological quirks: these need to be more explicitly enumerated.

    I find the passive voice generally harder to understand, less direct, and ugly. Not objective, but we could have a poll?.

    Comment by Martin Austwick — November 19, 2010 @ 9:19 am | Reply

  10. Yes, traditionally the passive voice is used to emphasise Descartes’s separation of the observer from the object of observation. So when written up in the passive voice, our experiments or observations seem to “do themselves” – we are not cited as a factor in the outcome, so anyone else repeating them should therefore obtain the same result etc.

    But these days most of us take that as a given in science. And the journal Nature, for example, now requires papers be written in both the active voice and first person.

    Daily Mail examples aside (and really, should that be our touchstone?!), the active voice is clearer, because a reader can follow exactly who did what.

    It is also livelier and more natural. I hope, for example, that you would not say to someone in conversation: “Eastenders was visualised last night [passive], and it was most entertaining”.

    I tend to retreat into the passive when I am not actually completely sure of what I want to convey, or details are being fudged – it can be a refuge for the vague.

    So when teaching students how to engage wider, non-technical audiences about their research, I give them George Orwell’s advice in his 1946 essay “Politics and the English Language” (note points iv and vi):

    (i) Never use a metaphor, simile, or other figure of speech which you are used to seeing in print.

    (ii) Never us a long word where a short one will do.

    (iii) If it is possible to cut a word out, always cut it out.

    (iv) Never use the passive where you can use the active.

    (v) Never use a foreign phrase, a scientific word, or a jargon word if you can think of an everyday English equivalent.

    (vi) Break any of these rules sooner than say anything outright barbarous.

    …and point them to his observation that “If you simplify your English, you are freed from the worst follies of orthodoxy… when you make a stupid remark its stupidity will be obvious, even to yourself.”

    For technical writing, what matters is clarity and precision. If the passive is sometimes required to achieve those goals, then so be it, rather echoing Orwell’s point (vi). But often you have to ask yourself “which is more important – conveying the specific point of this sentence most clearly by using the active voice, or keeping the passive to continue emphasising the general principle of Cartesian objectivity?”.

    I also like Strunk and White’s advice that we should choose our words, rather than letting them choose themselves. Thanks to the traditions of technical writing, I think that many of us fall into using the passive by default, rather than examining whether it is the most effective construction each time.

    Comment by Jon Copley — November 19, 2010 @ 10:10 am | Reply

  11. By the way – this discussion seems to have morphed into something else, bias and unconscious bias, in part. I hope that some of you that have taken the time to comment here might look at my last blog post on this very thing!

    Thanks to all for taking the time to comment

    Comment by sylviamclain — November 19, 2010 @ 10:10 am | Reply

  12. I am generally inclined to stay away from use of first person descriptors in scientific writing. However, that doesn’t mean that scientific writing need be passive! Many people interpret active vs. passive as first-person vs. third-person (or perhaps persons in absentia). You can make writing active without ever bringing people into it (unless, of course, you’re doing human studies 😉 ). You can turn proteins, diseases, processes, etc. into ‘characters’, so to speak. Instead of saying, “Reaction x is catalyzed by enzyme y” write “Enzyme y catalyzes reaction x”. This example isn’t so bad, but on many occasions use of the passive voice results in convoluted sentence constructions that are difficult to follow. Active voice allows you to tell a story with the subject of your experiments as characters. Plus it takes more words (and characters) to say the same thing, so learning to use active voice without first-person can save you some space when you’re writing abstracts or grants with strict character limitations.

    Comment by biochembelle — November 19, 2010 @ 12:35 pm | Reply

  13. […] subjectivity versus objectivity debate is illustrated quite nicely by the comment thread to Sylvia McLain’s recent post on active versus passive. The central assumption seems to be that using the active voice acknowledges the researcher as an […]

    Pingback by Active, Passive, Poppycock « Speech and Science — December 18, 2010 @ 9:26 am | Reply

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Blog at

%d bloggers like this: