Why it’s important to be exactly certain about how much you don’t know.

This blog post is another in a series towards developing ideas for the new book I am writing with my ANU colleague Shaun Lehmann and Katherine Firth of the Research Voodoo blog.

“Your academic writing trouble and how to fix it” was born of our frustration at reading the strange comments supervisors sometimes write on thesis drafts. Sometimes academic feedback makes even less sense than the text that inspired it. In this book we work backwards from these hard to interpret supervisor comments to tell you want we think your reader is complaining about how how to fix it. Writing Trouble will be a swiss army knife of a book, containing a range of strategies and tactics for fixing academic writing that is good, but not yet great.

Our publisher has been relaxed about us sharing our work in progress on our blogs and the process has really helped us make the book better – so thank you!. Parts of this post on hedging language will end up in chapter six: “Uncritical! How to make writing that persuades”. We are currently in the final (more boring) part of editing the book. This segment is the rough first draft I wrote some time ago. Katherine is currently working on expanding and polishing it up, so we welcome your feedback! If you’re interested in knowing more about the book before we publish, you can sign up for our writing trouble mailing list.

We are also collecting examples of hard to understand feedback to illustrate the book – if you’d like to share feedback you have received we are collecting them here. We hope to be able to offer people who donate text a discount on their purchase – stay tuned!

Sometimes writers get feedback encouraging them to be more assertive, like “I don’t hear your voice” or “yes, but what is your opinion?”. When you get this kind of feedback, your reader is frustrated because you are not ‘taking a stand’ in your writing. Taking a stand means making an argument for – or against – something.

While academic writers need to take a position in their writing, they must be careful not to over claim, especially when putting forward a theory to explain the observed evidence. Knowing exactly how much you know and don’t know about something – and how to write in a way that your reader understands your level of certainty – is fundamental to being an academic, which is why skilful academic writers know exactly how to employ ‘hedging language’. Being precise is one of the most, if not the most, important of the academic values and we must be as precise about our uncertainty as we are about everything else. Hedging language is tentative: words and phrases like might, maybe, sort of, I think, possibly and so on. These terms help us modify strong claims without losing valuable nuance.

Sometimes we see writing advice that suggests writers get rid of hedging language to avoid sounding ‘wishy washy’, but remember – when it comes to writing you are at a painful, middle-class dinner party. It is vital that you do not conflate ‘taking a stand’ with ‘writing forcefully’.

Just as yelling louder will not help you win a fight with your family member at a Sunday dinner, getting rid of hedging language to look more confident will not endear you to your academic reader. People not trained in academic ways of thinking can find the use of hedging language extremely frustrating to read. However, we are dealing with hard core notions of truth, and certainty, here – we must therefore be careful. With the possible exception of maths (in particular, maths proofs), all research is, to some extent, tentative. Hedging language introduces intentional vagueness to avoid sending clear signals to your reader.

You might be thinking – “but early in this book you told me I should avoid vagueness – now you are telling me to introduce it deliberately? What gives?!”. We know it sounds contradictory, but when you write academically you must bear in mind that you are communicating within academia, not just communicating about what you found out. Going with the idea that academic writing is a form of fencing or a passive agressive middle class dinner party, there are (largely hidden) rules around how you can express our interpretations of data.

Hedging helps us be intentionally vague so that the reader is forced to ‘read between the lines’ about what we think about a data. When we use hedging language we must balance between what we see in the data we are writing about, and the world from which we extracted the data. One motivation for doing this, according to Hyland (1998), is to “seek self-protection from negative consequences of poor judgment”.

Here’s a silly, totally made up example: imagine you have been taking photos of the night sky and have noticed there are more stars than there should be. You could form a theory that the extra stars are UFOs, but, if you want to be taken seriously in academic, you’d want to be very careful about how you write about this theory because, well – most people don’t believe in UFOs. The incautious student would write something like this:

“The extra stars shown in the table and images above are UFOs”

The word “are” signals that there is a direct correspondence between the data and the theory. It’s highly likely that an academic reader would just put a line through this sentence and write “rubbish”. If instead, you want to draw your reader into your theory while shielding yourself from ridicule you could write something like this:

“One possible interpretation of the data shown in the tables and images above is the existence of UFOs or other, unexpected stellar artifact”

This sentence is top notch hedging in action. We have actually said something faintly ridiculous (that UFOs exist), but we have left the reader unsure of whether we believe the statement either. We included a classic hedge word (possible) at the start of the sentence and then threw in a modifier (or) at the end.

In our UFO example, hedging language functions to distance ourselves from the proposition so we don’t entirely ‘own it’. It’s a bit like putting something on the table and backing away, then pointing at the item and asking the reader what they think it is, rather than telling them what you reckon. Hedges rely strongly on context to make sense – and reader will subjectively interpret them. In a strange way, hedging language helps us collaborate with our readers to find collective meaning.

The use of hedging language is connected to how we make knowledge within communities. Academia is profoundly hierarchical and it is important to bear this in mind when you want to disagree with anyone else’s theories, interpretations or evidence. We use hedging to signal to the reader that we are cautious, careful researchers who pay due attention to the accepted ideas and theories in our field, not cranks. Rarely is academic writing confrontational; the clever academic writer strives for an air of humbleness when they disagree with anyone.

Cultivating the right degree of academic humbleness is a matter of careful word choice, including careful use of hedges. This is one of the many reasons academic writing is accused of being obtuse, but it’s sadly unavoidable when you are a student and need someone to approve of your work. Hedging language is, therefore, a vital part of any researcher’s writing toolkit. Hedging language helps us indicate the precise degree of uncertainty we feel about a finding, fact or idea.

In the final book we will include a table and examples of how to hedge with style! Are there any questions? We’d love your feedback.

Related posts

Click here if you’d like to donate examples of feedback for our Writing Trouble Book

Don’t let those sticky words confuse your examiners

Academia is a painful, upper middle class dinner party

Love the Thesis whisperer and want it to continue? Consider becoming a $1 a month Patreon and get special, Patreon only, extra Thesiswhisperer content every two weeks!

10 thoughts on “Why it’s important to be exactly certain about how much you don’t know.

  1. I’m not completely convinced by this piece. I think it’s because it presents a single solution (writing using hedging language) to the problem raised (making controversial claims without making it clear you know they’re controversial will get you in lots of trouble with reviewers, examiners, supervisors etc). I’m convinced that the latter is a problem, but not that the solution presented is the only one.

    It reminds me of a discussion I once had with one of my supervisors. I’m doing a science education phd, and my supervisor’s background is in the science discipline, not qualitative research. I had shown him a quote I intended to discuss in my thesis and presented my interpretation. He seemed a bit concerned and eventually commented that he wasn’t convinced that this was the only way the quote could be interpreted, but supposed that was ok because in qualitative research lots of interpretations of the same quote are possible and I should mention this when I wrote it up.

    Instead of immediately following this suggestion, I asked him if there was a particular aspect of my analysis that he found unconvincing. In the resulting discussion, I was able to explain which parts of the quote led to each component of my analysis. By the end, he was convinced by my interpretation and changed his advice to saying that the write-up needed to include the sort of explanation I’d just given him.

    Going back to your book section, I think that if we get feedback that the theory we’re using to explain data is ‘unconvincing’ or (ouch!) ‘rubbish’, we should consider whether:
    – the claims we’re making are too strong, which can be addressing by hedging language;
    – we’ve missed out key parts of the argument that justifies the claims; or
    – we need to go back to the drawing board and come up with a better interpretation.

    • Thanks for your feedback Ruth – we do actually tackle a lot of these issues throughout the book, so once the post is in context I think your points would be addressed with references out to other sections. Your last three points at the end are useful clarifications though, so thank you!

      • True, one of the main advantages of your book is using the cross references to work out what underlying issue a particular comment indicates. Without them (and the rest of the book) it’s hard to appreciate the context of an individual section.

        That said, I think some of my issues may be coming from the choice of example. Taking my three scenarios, I could rewrite 1& 3 as:
        -the claims are plausible and interesting, but not the only position that could be taken -> hedge
        -the claims are implausible or based on very weak evidence -> back to drawing board.

        Your UFO example takes a ridiculous claim and tries to hedge it, rather than replacing it with something more sensible. The end result is a reader wondering whether the author really believes the nonsense they’re writing, which is not the sort of vagueness I’m aiming for when I write.

        If your examples were based around more plausible claims, I think they would better illustrate the point in the title, about being clear about what you don’t know. For example:
        “The unique approach of Writing trouble will solve all your writing problems and mean you’ll never have to read another book on writing again”
        “The unique approach of Writing trouble is expected to effectively complement other sources of assistance to help you solve your writing problems.”
        In this case the vagueness introduced through ‘expected to’ maps directly to something we don’t know: how readers of a currently incomplete book will respond to it.

  2. Nice work! I enjoyed the read, and, while I appreciate the validity of Ruth’s comment above, I was thinking that perhaps the other issues raised have been covered in other chapters of the book.

    This post however put me in mind of Michael Quinion, he of OED fame. He wrote briefly about where the word hedge comes from in 2006 “which could at one time mean any sort of enclosing barrier, not necessarily a row of bushes or small trees; Cornish hedges are stone walls, over time copiously obscured by vegetation, a fact occasionally discovered too late by tourists trying to leave room for another car to pass in a narrow road by driving into the ‘hedge’.”

    Hedges don’t have to be wishy-washy 😀

    Quinion, M. (13 May 2006). 1. Feedback, Notes and Comments: Linhay. World Wide Words, 487. Retrieved from http://listserv.linguistlist.org/pipermail/worldwidewords/2006-May/000378.html

    • I agree, klinkehoffen!

      The goal of “hedging” language is to be specific about the boundary between what we know and what we don’t know. If done correctly, there is nothing wishy-washy about it. However, as with all writing, there is the possibility that hedging could become wishy-washy if the writer really isn’t sure where to place the boundary between certainty and conjecture. For example, what if the writer of the UFO paper were to say:

      “Our photo of the night sky showed both objects that have been classified as stars (Famous Astronomer, 2006) and others that do not appear to have been classified as stars. We therefore postulate that it is possible that these objects are indeed not stars. According to Infamous Astronomer (2007) there are UFOs in the galaxy. If the objects we saw were not stars, then it can be conceived that they are other non-stellar objects such as UFOs.”

      We would be left wondering if the writer was unsure about their interpretation of Famous astronomer’s classifications. And, we would probably stop reading before we reached the end of the paragraph!

      Coming from a non-academic background, I must say that I find a lot of academic writing to be frustratingly supercilious and long-winded. However, I don’t mind reading a few hedging words here or there if they serve to clarify the meaning.

  3. On a somewhat tangentially related point, one piece of advice I was given is that as you approach the end of your research and start writing up your thesis, you have become one of the leading experts in the narrow field of your “significant contribution”, so you can expect that the examiners along with future readers might need to be “taught”.

    The consequence of this advice, for me, was to write some chapters in an exposition style. Very much along the lines as if I was delivering a teaching lecture on that very topic.

  4. Pingback: Academic Small Talk - Café cum laude

  5. Pingback: Your Thesis – Why it’s important to be exactly certain about how much you don’t know | Learning Change

What do you think?

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

WordPress.com Logo

You are commenting using your WordPress.com 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