What should it be like when it is done?

We are busy again with drafts for the forthcoming calls  and so we have to be able to agree with the researchers when the job has been done, when to call a halt to the work whether this is at the end of a long dialogue or (on rare occasions) after one quick pass to put things in order.  Like writing work, in the final analysis it is more a question of abandoning the work rather than polishing it to perfection as one tends never to be entirely happy to step away from it and send it out to the world. However, when the work is in a very advanced state of readiness and is prepared to withstand the most critical, or (more importantly) the laziest of readings it tends to have the following characteristics to which all proposals need to aspire and towards which we always push them.

In this post I’ll just run through quickly the different aspects of the type of reading experience that the writer should be aiming to supply the reader with in order to make the work easy to evaluate and optimise the selling aspects of the game that we are playing here.

Texts that are submitted with the following characteristics are tough to deal with and tricky to evaluate regardless of the quality of the material that might be in there often hidden deep under layers of opaque and confusing sentence piled on sentence.  There are others which I’ll try to cover in future posts, but for the moment we’ll take a look at three.

Ideas and meanings proliferate – the basic terms of the argument are never quite clarified (this is a common comment in the evaluation reports, more common than you’d imagine, in fact) and shift slightly at different stage of the text and the set of ideas grows as the texts are built up layer on layer of ideas and activities rather than carefully and ruthlessly pared down as they must be to elegant and spare structures where only that which is necessary and sufficient to reach the objectives ever makes it to the page.  Ideas often come in cascades, one clearly bringing another to mind which prompts  another and so on. We have dealt recently in another post about the difficulty of building a project from the bottom up like this allowing good ideas to propagate and grow in search of an objective to explain their importance and relevance to the state-of-the-art.  It is much quicker and more effective to set out the objectives and draw in the ideas that are needed to reach them, to let the crystal clear objective statements pull the whole edifice into being from the way in which the state-of-the-art is interpreted, through the deliverables and methods to milestones and activities.  Start at the end and work backwards is really the only quick way to do it – and this applies equally to other parts of H2020 and other proposal writing tasks.

Makes its own laws – the basic structure of the proposal is clearly set out in B2 i.e., there are a set of headings that should also be used to structure the B1 for speed and simplicity.  Too many researchers overlook this structure and the logical flow that it forces them to use to present their ideas (we have seen in other posts that the structure of the proposal reflects a deeply held view of what science is, in fact, in these programmes and explains many key ideas including what is meant by new and risky ideas, see recent posts for more on this).  There is a particular ERC vocabulary (frontiers, individual researcher-driven science, basic science, etc. etc.) and a particular set of structuring concepts (objectives-driven research, risky, over the edge of the state-of-the-art research, ‘pioneering’ individuals rather than teams etc. etc.) that we can’t avoid, in my view, if we are going to be competitive in the short amount of time we will be allocated during reading to make a strong impression.  And yet, many researchers still invent structures (in particular in the B1) which barely touch on the ones that are expected to read.

The reader becomes the writer – this is probably the most important and common problem that I see while reviewing draft ERC proposals.  Many are so loosely put together that the reader is forced, if they are to make any sense of it at all, to put the pieces of the puzzle in some kind of order and then read it for the purposes of the evaluation.  This, as I have written in a number of previous posts, is simply too big a risk to take – they evaluators simply don’t have the time to do this work and are skim reading rather than carefully linking the parts before then reading to assess the quality of the ideas.  In practice if the evaluation comments are anything to go by then the evaluators often run out of steam and struggle to find anything in particular not to like other than a sense that they don’t really know what it is supposed to be about.  They write something vague about it not being very interesting and move on to the next one.  I’d say, that above everything else, it is imperative that the reader is not burdened with this task of making the text cohere.  The work must have an unmistakeable thread running through it that carries the ideas and carries the reader along with it so that the work needed to get to the ideas is minimised.  A vast amount of the work that we do on proposals is trying to open the text up to reading and assessment and texts that demand too much of the reader are very common indeed.

In stark contrast, well polished ERC proposals have characteristics that are the contrary to those set out above.  These in my view are the things you should aim at to give yourself the best chance in this very tough competition.

Familiar, anticipated form and linear structure – the evaluators have a certain simple structure in mind when the sit down to read these texts (with a very large drink one can only imagine) i.e., basically,  introduction, state-of-the-art, objectives set in context of state-of-the-art, results, methods to be used and resources and not very much else. If you give them this very obviously and systematically in a drip feed kind of a way where one thing is set out, the next introduced and done too and so on they will be very grateful and look kindly on the work.  But it is strangely rare to find this simple and linear presentation and many texts jump around from one part to another or one category of idea to another and the reader has to work: the last thing we want is the reader to have to work, not even really to think, and certain not to find it a labour.  The structure is there for you, I’d consider very seriously using it to make it easier to read, to make it recognisably and ERC project and to make it much easier to read and make it ‘evaluatable’, if such a word exists.

Predetermined meanings, fixed meanings, closed text – don’t assume that the reader who wearily picks up this text after a heavy lunch is going to know what the terms of the debate you are creating actually are.  I have seen comments quite often which say that the terms were not spelt out and that they seemed to shift as they were used at different points of the text.  Take time to say what you mean by each term and idea in the context in which you are employing them – as far as possible there should be absolute clarity on the vocabulary.  The reader to aim for is, and the is I have written about elsewhere, is an intelligent lay person, a non-expert in the early phases of the work where it is all set up before ploughing into the fine details of the argument and the methods in the latter sections where this more developed argument is necessary and can’t be avoided.  But on the whole, we should aim to use shared language and the concepts they capture.  The text is also made by removing ideas rather than adding more and more.  The end result is spartan and spare with no excess and with no ‘interesting’ lines of investigation opened or unanswered questions posed.  The logic tends to be rigorous and easy to follow up and down the proposal i.e., one level is caused by that below and leads to the one above and all in the end leading up to the breakthroughs captured in the objectives statements – it is all to do with utility, nothing to do with ornament: unfortunately, too many proposals contain material that is neither use nor ornament.

The reader is passive – in contrast to the proposals where the reader has to do lots of work to make the text coherent enough to attempt an evaluation (a very risky strategy indeed), the reader of a very well polished text hardly has to read at all. In fact, in a deep sense, really critical reading is something we’d ideally like to avoid them doing.  If we can make something so smooth and persuasive and logical that they don’t need to engage the critical mind (which is where the fault finding will also take place) then we are getting towards the end of the process.  Style and structure play a role in giving the reader a work-free experience as we don’t want to jolt them into wakefulness by making it tough or boring to read – so plenty of sub-headings and space for the text to breath and no page-length unbroken paragraphs.  And the rest is really about linking the parts and the different types of argument that needs to be used at different phases of the work (I have written elsewhere in this blog about the difference between the inductive logic of the first sections and the deductive logic of the latter sections and how to link them up).  Simplify, contextualise, commit and link it all up and the reader will be so grateful that you’ll already be head and shoulders above similar quality science stuck inside an ugly and lumpy package that is impossible to unpack.