Do you really believe what you are writing?



I often make my doctoral students cry, but I hasten to add it’s not because I am mean. The supervision work I do is emotionally intense because I seem to have (accidentally) become a specialist in helping people who have had difficult candidatures for one reason or another. Gina Wisker calls these people ‘doctoral orphans – which is a nice description I think.

Doctoral orphans are interesting people. Wisker claims those who finish have become confident self learners. Certainly the ones I have met are self reliant and clever. Many academics, however, shy away from taking them on under the impression they will be too much trouble.

This is a pity because often orphans have developed advanced coping strategies and are usually fast learners once you point them in the right direction. One technique is to take the existing writing apart and analyse it together (for a good explanation refer to Kamler and Thomson’s ‘Helping Doctoral students Write‘).

But taking your writing apart and properly analysing it is a bit like renovating a house. What you thought was a simple job can suddenly become complex. You can find all sorts of stuff was not done properly in the first place. You might find that some problems are too expensive to fix and you are forced to knock it over and rebuild.

Recently I managed to provoke a full blown existential crisis in a student by using this co-writing method. As we started taking his writing apart I realised I couldn’t find any warrants. Warrants are the most difficult part of the standard argument ‘kit’ to understand, let alone explain. I did a particularly crappy job explaining the problem (in my defence, I was horribly jetlagged). Since #acwrimo starts tomorrow (get your nerd on people!) it seemed appropriate to tackle this topic in more detail.

We all know that when you make an argument you must provide reasons to back it up. Your reader may not accept your argument if they don’t believe the reasons you give are relevant. This is where warrants come in. A warrant is there to convince the reader that the reasons for your argument are valid.

It’s easier to understand how warrants work through examples. Let’s imagine for a moment I have done some research on the topic of doctoral orphans. I could start to make an argument about them with a statement like this:

“Doctoral orphans are often better students than academics think they are”

So far so good. I’m making a ‘knowledge claim’ about orphan doctoral students, but why should you believe me?

I need to give you a reason, like so:

“Doctoral orphans are often better students than academics think they are because most academics do not have direct experience of working with them

Some people would be perfectly happy to accept such a statement, but I can’t be SURE. A tricky reader might for instance ask:

“Why does an academic necessarily need direct experience to know something? Can’t they learn from someone else?”

I can shore up the reason for my argument by providing a warrant. A warrant is a general principle and can take a number of forms according to Booth, Colomb and Williams in their excellent book, ‘The Craft of Research’:

  • cause and effect
  • one thing is a sign of another
  • a rule of behaviour
  • a principle of reasoning

Let’s try a ‘rule of behaviour’ warrant, like so:

“Doctoral orphans are often better students than academics think they are because most academics do not have direct experience of working with them. Hearsay is not a good way to find out about a student cohort“

That’s ok, but it’s a bit thin and awkward. Let’s it again in cause and effect mode, this time making the warrant have two parts: a general circumstance and a general consequence.

“Doctoral orphans are often better students than academics think they are because most academics do not have direct experience of working with them. Without direct experience are forced to rely on rumour and ‘what everyone knows’ – which might be wrong.”

That’s better.

But do I even need that warrant?

This is where it gets complicated.

A warrant should be, according to Booth et al, a ‘common sense statement about the world that everyone considers self evident’. All disciplinary communities have their own forms of ‘common sense’. Sometimes a warrant can be construed as condescending because you are telling the reader something they already know all too well.

Booth et al give this good, simple example of an extraneous warrant:

“Don’t walk down the stairs at night because you might trip over. It’s easier to trip over when it’s dark“

Most adults know that it’s easy to trip when it’s dark – they don’t have to be told. Booth and his co-writers point out that we only need to provide a warrant like this only when speaking to children. We don’t want our readers – particularly our examiners – to think we are treating them like children.

This is why leaving out the warrant can make an argument stronger. Are you confused yet? Stay with me!

Remember, academic writing is terribly polite, even passive agressive. Leaving out a warrant is a subtle signal to the reader that you are ‘one of us’ – part of a knowing community. If we decide most people reading the piece will consider the warrant unproblemmatic, we can proceed straight from reason to our evidence and then a closing statement, like so:

“Doctoral orphans are often better students than academics think they are because most academics do not have direct experience of working with them. Some 5% of doctoral candidates experience a change in supervision, but many persist in relationships which are dysfunctional (Mewburn, 2013). Some candidates can ‘fall through the cracks’ and remain effectively unsupervised for long periods of time. When they eventually do seek help, our research shows, the ‘bad reputation’ of orphaned students, formed out of stories circulating amongst colleagues, can affect individuals adversely as they seek alternative supervision. If academics had better information on the challenges and opportunities of working with this cohort, this ‘bad reputation’ could be overcome.”

That looks pretty good to me. I don’t think the warrant was adding much, but you might disagree.

Warrants highlight one of the main struggles of academic writing for the novice (or for experienced writers entering a new area for that matter). Knowing when to leave out the warrant is like learning how to season the writing stew. You develop a feel through immersion in a knowledge community, you can’t learn it in advance, or all at once.

I might add, even seasoned professionals like myself find warrants tricky. I had to consult two of the best writers on writing I know – Pat Thomson (‘Patter’ blog) and Rachael Cayley (‘Explorations in style’ blog) – while writing this post to see if it made sense (any remaining errors are my own of course!). Pat agreed it was hard to do in a blog post and recommended this page of exercises. Rachael remarked in email:

“I feel like I see a lot more unstated warrants than I do explicit warrants; I often find myself asking the same question: “Can this move go unexplained in your field?” “Are you sure?” The resulting conversation is usually enlightening because the student realizes their growing disciplinary expertise (or the areas in which they are still feeling their way). “

This remark echoed the existential crisis experienced by my student. When I asked the questions like: “how do you know that?” he realised he didn’t have a warrant and couldn’t think of one. He then started to wonder if he believed in some of the things he was writing. To his credit, he faced this bravely and it sparked a fruitful discussion about possible directions the thesis might go instead.

It’s going to be a bigger renovation job than we thought.