The academic ‘we’

Once there used to be a royal ‘we’ – the word ‘we’ used to mean ‘I’, as in Queen Victoria’s ‘we are not amused’. These days royals almost never say ‘we’ unless they mean ‘we’. The current British monarch doesn’t even always use ‘we’ when she could. In a phrase association game, the answer to ‘my husband and I’ is ‘Queen Elizabeth II’.

Yet there is one place in which people still say ‘we’ when they mean ‘I’, and that is academia. Only a couple of weeks ago I had to remove the academic ‘we’ from a sole-author article to be published in Policy. At his blog, Andrew Leigh labels this usage ‘pretentious’. Insofar as the academic we is an implicit reference back to the royal we, he is right. But I am not sure that is what academics are consciously doing when they write ‘we’ instead of ‘I’.

Perhaps it reflects the collegial tradition within academia. As Damien Eldridge (himself an academic) writes in Andrew L’s comments:

even if the paper has a single author, it is quite possible to interpret the we as referring to the author and the reader.

But this saves ‘we’ from being pretentious by making it presumptuous instead. If a solo author writes in an academic paper ‘We conclude that…’ or ‘We believe that…’ my response is ‘speak for yourself’. The writer is trying to persuade the reader; the reader decides whether the argument warrants a ‘we’.

The ‘I/we’ problem can be solved by removing personal pronouns entirely. Over the years my articles have lost a few uses of ‘I’, excised by editors who insist it has no place in academic writing. While good stylists can keep the ‘I’ count low, a personal pronoun prohibition doesn’t always work well. If an author wants to draw a conclusion, but at the same time leave open the possibility that others could reasonably disagree, expressions like ‘I think’ or ‘I believe’ are preferable to clunky formulations like ‘it could be argued that…’ or ‘it is suggested that…’ , as if arguments mysteriously exist without anyone in particular making them.

‘We’ adds authors and ‘it’ subtracts them. With one author, the precise and concise ‘I’ is better than either.

15 thoughts on “The academic ‘we’

  1. Last month as an associate editor of an international technical journal I attended a meeting of the editorial board. For the record, the journal is not entirely academic, industry types represent more than half the board. This very topic was thoroughly discussed. One of the editors complained that the copy editor edited out all the ‘we’ from his own paper. It was clear from the discussion that some Americans in particular regard ‘we’ as royal and pretentious, while many others feel that “I” is full of self-importance. Apparently, the editorial office checked some authoritative sources (like Chicago manual of style) and concluded that both are permissible. Some suggested that we as ‘I and the reader’ is more than appropriate in some context. You are right about ‘we conclude’. But what about ‘substituting a for b, we obtain’. If I say ‘I obtain’, it may mean that someone else may obtain a different result.

    In Russian ‘I’ would be extremely arrogant.

    Eliminating personal pronouns is a non-starter. Good English encourages active voice.


  2. “Good English encourages active voice.”

    But how often is active voice found in academic journals?

    I can’t see that ‘I’ is self-important unless the author is unnecessarily talking about himself/herself. And if it is self-important, why don’t we have anonymous articles?


  3. I think Andrew Leigh and Damien Eldridge are both wrong about this.

    “We” in formal writing usually seems to be a well meaning but amateurish attempt to avoid the appearance of arrogance. By using this construct, the writer is trying to make his or her assertions less prominent.

    Sometimes this is done precisely because the writer is indeed arrogant, and possibly Andrew L has come across a few examples in that genre. Simply using “we” does not disguise arrogance.


  4. I think “we” in academic publishing comes from the tradition where a scientific paper is thought of as an objective piece of research, as opposite to an opinion piece in a newspaper. Thus it has two functions: to involve the reader (as I do routinely during my lectures when taking students through a math derivation. We do it really together. Same in the paper). The second meaning is to use “we” to make it impersonal and objective, AND use the active voice.

    Ultimately, it is a matter of tradition. Language is a living thing and has different traditions in different areas of knowdlegge. What’s wrong with that?


  5. I always took “we” to mean the author(s) and the reader – the author(s) are leading the reader through whatever discussion and argument they are making. There’s also the sense of self-importance in using “I”.


  6. “I” is not uncommon in sociology, particularly ethnography and derivative fields where the existence of the researcher’s subjective point-of-view is unquestioned.

    Recently I used “we” in a paper to mean “my co-researchers and I” — shocking, I know, to use a common English word in an academic paper in the normal way!


  7. The rule banning “I” from academic writing is premised on the delusion of pristine objectivity of the writer. I have recently taken to using “I.” It is so liberating.


  8. I’ve just surveyed a recent paper where I was a sole author (a rare case for me). I found that I have “I” four times and “we” 9 times. In some of these cases I could (and probably should) have used “I”. But in some cases it would look funny. Especially a construction “in this case we have a=b”.

    Maybe it depends on the field. I am sure math is more objective than sociology.


  9. Boris – I think when a fact is indisputable common knowledge, ‘we’ is acceptable. But where it is your own finding or conclusion ‘I’ is better.


  10. Andrew, one thing that neither of us mentioned is the influence of which side of the Atlantic influences you the most. I think Americaphiles are far more comfortable using ‘I’, while Anglophiles tend naturally to ‘we’. (I’d be curious about how places like China and Japan handle this.)


  11. Over in Korea, which is probably the most erudite culture in the world, having produced the first printed books (not Guttenberg), and where parents spend up to 50% of their income upon education, they say ” it is ” rather than I or we when concluding.

    For example, “yepo imnida” means “beautiful it is” (polite form). The contextualisation is up to the readerly interpretation of what is presented by the author rather than ideas being given personal pronouns. This occurs in Pacific Englishee too..


  12. Off-topic, but I find that even some academics refer to ‘data’ in the singular, such as, “the data shows…”. Of course, this is a lot more common outside academe.


  13. Rajat – And I think they are often right to do so. I think it is silly to import Latin grammar into English, and people don’t do so with similar words. How many people would say ‘The agenda say that the next item is…’? The ever-sensible Pam Peters advises using the English distinction beween mass nouns and count nouns. Count nouns, as the name suggests, refer to things that can be counted. ‘Persons’, ‘blogs’ etc. They take the plural. Mass nouns refer to concepts with less clear boundaries, ‘information’, ‘mud’. They take the singular.

    ‘Data’ can be either, depending on context. For example, if I said ‘Andrew Leigh uses a lot of data in his research. This data is interesting.’ The singular is correct in this case, because I am referring to an undifferentiated collection of data sets. If I was referring to particular set of data it would be a count noun and take a plural verb.

    However, in spoken English ‘data’ is mostly singular.


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 )

Google photo

You are commenting using your Google 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 )

Connecting to %s