.post-body { line-height:100%; } -->

Friday 4 February 2022


We've all done it, I'm sure; leapt out of the bath and run off naked down the street, having solved one of the thorniest problems in philosophy.

OK, maybe we've not all literally done it in the way Archimedes is purported to have done in what's almost certainly an apocryphal tale. Still, over the years, I've encountered many, many examples of people doing the easy bits of Feynman's famous problem-solving algorithm and running off metaphorically to expose their pendulous danglies to the internet at large without doing what Archimedes had done prior to his extra-ablutive excursion and fulfilled the difficult middle step.

Just in case anybody's still unfamiliar with Feynman's thoughts on problem-solving, he can be summarised in three steps:

1. Write down the problem.
2. Think very hard.
3. Write down the solution.

Of course, Feynman, inveterate scamp and fan of nudity though he was*, and having solved many problems considerably more difficult than fluid displacement, never felt it necessitated streaking. Nor, I suspect, did Archimedes. 

As we can see, the middle bit is where the wheels almost always fall off. Or is it? Today, I want to take a little break from thinking about thinking to write about writing. Or presentation, more accurately. This is a quick and dirty offering on how to get your idea taken seriously.

In the last couple of days, I've encountered a new example of a very familiar story, but it's such an egregious example of a very particular problem that I've encountered many, many times that it seems to warrant a wander into how I think about writing when I have ideas I want to express. It's not intended as a comprehensive style guide, just a basic pass at how to present ideas.

Many's the genius buried a fool for lack of the right words, they say. In fact, they don't say this, I just made it up. That doesn't diminish it, though (nor the corollary statement that many a fool has been buried a genius because they had a vocabulary).

How many good ideas have been lost because of incoherent presentation, or rambling dissertations of mostly irrelevant biography or ideology that serve only to obscure whatever it is that's being presented?

I won't present the latest example of this for debunking. There are lots of examples of bad ideas being debunked littered around the blog. In fact, the reason this example has motivated me to write this is that, in a video presentation almost an hour long (which I did not watch; I read the maker's own transcript having checked the first few minutes for accuracy) I could discern no clear statement of the idea. Attempts to clarify with the author have met with hostility and the usual deflections about our competence; this on a small internet forum populated with almost entirely science writers and creators, all well-experienced in reading dense technical literature for conveyance to a lay audience.

This is a real problem. Of course, writing is a skill, and writing for a particular audience or purpose is a different skill. That's right, they're not the same. Writing in very specific ways requires learning those ways, and behaviours and conventions we learn are skills. Regular readers will recall, for example, how incredibly different my entire style of writing was when writing about a personal experience of a Renaissance masterpiece than here, where I talk about deeply technical subjects. Precision is important in this context, emotion more so in the other.

So here's my approach to presenting ideas coherently in one easy post. It's not exhaustive, and it's by no means meant to be an academic style guide, it's really just a guide to legibility so that, if you really do have a good idea, we don't all just dismiss it as idiocy because it was poorly or ambiguously expressed.

Now, the exact nature of your presentation is going to depend entirely on your target audience. In particular, if you're targeting gullible rubes on the internet, you'll present in one kind of language (mostly catchphrases, slogans and soundbites for the short of attention). If you're presenting a defence of your doctoral thesis on entropy in superfluids, you'll present in another kind of language (differential equations). It's a spectrum, with lots of stopping places possible on the way. Whether you're aiming at wowing gullible rubes with your deep insight or you're presenting a defence of your doctoral thesis, however, the general guidelines for presentation here will serve as a good guide.

First, some general principles:

1. Be concise.

It can't be overstated how important this is. It's tempting to provide every word of motivation you can think of. I get it. Everybody wants to tell their story. From your perspective, your idea is deeply intertwined with you, and everything you've felt and thought in arriving at your idea is of enormous relevance to the idea, but this is your bias. In reality, the idea stands or falls on its own merits. 

2. Stay focused.

Digressions are fine (regulars here are more than familiar with my penchant for Tolkienesque rabbit-holes and thousand-word footnotes), but keep a narrow path. In academic presentations, digressions should always be avoided.

3. Draw a line.

This is true of all writing. Look at how I began this post. A conversational allusion to a well-known story from the history of science to set a starting point and the tone (coupled with the title), then straight into an expression of the problem I purport to solve (incoherent presentation), and the solution (better presentation), and now I'm drawing a line from that point through a full explanation of my solution.

It needn't be strictly linear, but there should always be a consistent anchoring idea that threads its way through your presentation to keep your audience focused.

4. Know your audience.

This shouldn't need saying, really, but it does. How you communicate ideas to different audience is a massive factor in how seriously they'll be taken. One thing in particular stands out; egregious sesquipedalianism. That might sound ironic, coming from among the most sesquipedalian of us, but the qualifier is important. Don't use long words because they're long, use the right words for the job. That also means having some grasp of the relevant material and terms of art (jargon) used in the field. Make sure your usage is either consistent with the general usage in the field or give your definition and your justification. This also applies to co-opting terms from elsewhere. Think of the language as a toolkit, and put in what you need for the job at hand.

Let's move on to specifics.

The most important single constituent of your presentation is the idea itself. Everything else you write needs to be cast in direct context with the idea. That gives us a nice starting point.

1. Start with the idea.

For me, this is the most critical part of any writing endeavour. In fact, because of my personal history as a writer, primarily songs, I don't put pen to paper until I have a title. Even now, everything hangs on the title.

There are lots of rules about titles. Many of them are bollocks, or based on commercial impact criteria like search engine optimisation, but the title should be an expression of the central idea. My title here, for instance, is an oblique reference to going off half-cocked (or cock and spuds doing a passable impersonation of a Newton's cradle) and ill-prepared, with the appropriate connections made in the first paragraph. For an academic paper, something that more directly expresses the idea or the problem, like On the Electrodynamics of Moving Bodies, to cite a favourite hereabouts.

Although it's not as obvious as it would be in, say, a scientific abstract, my preamble here is very much exactly what you'd find in such an abstract in content terms, although I've expressed it by prosaic analogy, allusion and other literary devices; I've stated the problem and presented a candidate solution.

A concise statement of the idea as early as possible is absolutely the most important part of your presentation, whether by direct expression or the problem it purports to solve. Failing to do this will result in the ignominy of deaf ears. There are all sorts of reasons for this, but the most pertinent is that it gives the listener/reader the context for every statement in your presentation. It can, in fact, mitigate a lot of the woolliness inherent in most presentations simply by providing that context. The more I have to read before getting to any semblance of what the idea is, the more probable that I'll simply have switched off or dismissed you as yet another crank.

In fact, your statement of your idea should be a limit; a constraint; a clear and unambiguous statement that can be tested against the relevant metric.

2. Meet the field on its own terms.

This is hard to hear, but it's necessary. The jargon isn't just words. Sure, it's incredibly irritating when somebody talks to people outside their field in jargon, and it's easy to get lost in it. However, if you propose to add to the sum of knowledge in a given field, you need to understand the field which, of necessity, requires understanding the jargon of the field. Terms of art are necessary in specific domains because they remove ambiguity and save entire regions of forest from the preposterous amounts of paper we'd have to expend in explanation. More importantly, if your ideas pertain to a field but you can't express them accurately and correctly in the language of the field, your idea will be consigned to the dustbin of intellectual history without ever getting an airing. 

It's fine to talk about a field in natural language, of course. It is important to take care not to stray too far from robust terminology and definitions, though, because that increases ambiguity.

Ultimately, if you have an idea that pertains to a particular field and you haven't done the necessary study to understand that field, you're unlikely to make headway, not least because the probability that you'll understand the objections of real experts - or even recognise real expertise in the field - is very low.

3. Present data.

There are very few data about the world that aren't fairly straightforward to acquire. If there are any quantities in your idea that pertain to the world, it's generally easy to find the stats. If there are scientific papers supporting your idea, present them. Don't just say in your words what you think they say, say what you think and then quote and cite the text of the paper so that others can determine for themselves whether your assessment is accurate enough to support your conclusions. Again, scientific papers are generally fairly easy to acquire. Even most paywalled papers are generally available after a year or so. If you need access to new stuff, find a friend with institutional access and ask if they can get it for you. If your scientific support for an idea is a news article, you have no support. You need track the news back to the primary scientific literature and make sure it says what the news article says it does. Newspaper writers more often get science horribly wrong than right, with only specialist science journalists maintaining the requisite standard. 

4. Address objections.

This is one of the hardest to hear. You're going to come across people who have objections to your idea. It's inevitable. It's not necessarily the case that all their objections will be valid, but you need to be able to assess them and deal with them, either by showing their irrelevance or by showing how your idea circumvents them. Again, ideas stand or fall on their own merits. This also goes back to point 2 above, because the ability to assess objections for validity will require an understanding of the language in which experts will present their objections. Failure to address these two critical points will get you assigned to the Pauli bin**.

5. Do the math.

OK, so a common objection is that explanations can be done without math, and that's true enough, as anybody who's read more than one or two posts here are more than aware. However, in order to show that your idea is effective in explaining some aspect of some phenomenon, you have to be able to quantify its contribution so that it can be measured. No measurement = no idea.

6. Check your crackpot score.

This is slightly tongue in cheek, but there's a famous list of the sorts of thing that crackpots include in their writings when presenting their revolutionary over-unit energy production device, or their groundbreaking idea of how the whole universe is subject to Darwinian-style natural selection. Physicist John Baez produced it, and it can be found below. It's a kind of mirror-analogue of this post, a how-to of how not to present scientific ideas.

If you have a brilliant idea, don't let it get lost by poor presentation. Make sure your cods are properly cradled prior to embarking on public excursions, and that you've properly prepared for what awaits you.

It's the only way you'll ever find it.

Further reading:

Crackpot Index - Score chart for whacky science presentations.

The Alternative Science Respectability Checklist - A wonderful blog post by physicist Sean Carroll related to this topic.

In on the Secret - Jargon and specialist notation and why we shouldn't fear them.

Very Able - Expertise and understanding variables.

dO yOuR rEsEaRcH!!! - How to do proper research and some of the pitfalls of poor research practices.

Well, Blow Me Down! - Why it can be worth persevering even when some experts say it's wrong.

*Feynman famously played bongos in a strip club.

** The Pauli bin is where we dispose of ideas that are 'not even wrong'.

No comments:

Post a Comment

Note: only a member of this blog may post a comment.