Tuesday, 9 June 2009

Problems With Logic

No matter where I look on the internet lately, I find people making complete arses of themselves.

I recently reviewed a book in which the writer had based his entire premise on a data-set which didn't actually prove his hypothesis. Instead of wondering if the lack of supporting data indicated that his hypothesis might be flawed he worked with the 25% of the records which did support it and ignored the rest. Now, I can see the problem here; you can see the problem here... poor bloke. A whole book.

YouWriteOn's ill-advised publishing scheme provided several excellent examples of odd logic with (among other things) its insistence that all rights would be returned if the writers chose to cancel. It was kind of them to make this offer: but it’s nonsensical, because you can’t do the same thing more than once and consider it a first time each time; so they couldn’t return first rights to any of the books that they published no matter what they promised.

The discussion which followed this blog post over at the Guardian relied very heavily on sweeping generalisations, opinion presented as fact, and assumptions based on untested or pick-and-mix data; and it spilled over in a little way to a post on this blog, where a well-meaning self-published writer tried to support her argument with facts that didn't actually prove what she thought they did.

I could go on: I have hundreds more examples, but I think you've got my point.

I am so tired of watching these lapses ooze across the internet that I have decided to write the occasional post here about logic, fallacies, and proper research.

What has this got to do a with publishing? A whole lot. When a writer relies on fallacies to make his point, and cannot support his assertions with solid research, he not only makes himself look foolish: he ensures he’ll never get his work published by any reputable press.


csmith said...

Well said. And the problem is that the research should be complete across the board. For instance, as an architect, nothing throws me faster than something wrong architecturallly - even though I know a good 80% of the readers won't pick up on it. And I'm sure that is true for other disciplines!

BuffySquirrel said...

Trust the good old Grauniad to miss the point; the agents were trying to help authors, not belittle them. The same cannot be said of much of the converse.


I often get previously-published submissions in slush with the comment that the author owns "all rights". Umm, no. You can't 'unpublish' something.

Trouble is, people who want to believe x about publishing will gravitate to those sites that validate their beliefs. That's human nature!

Dan Holloway said...

If that were the opening line of a novel, you'd have a smash hit on your hands :-)

I have a nostalgic fondness for that Guardian post - it was the first place I really came across Jane and this blog other than tangentially. I disagreed (I still do) with Jane on a large proportion of topics (looking back - it's strange that the actual article - left behind in the waves of generality - was on the subject of #queryfail - which, for all writers, should be sufficient reason for joining twitter) but I was sturck by the fact that I found myself getting hot under the collar more with the people who shared my opinions than with her. There's nothing more damaging to your cause than well-meaning people who argue it badly (FAR more damaging than people who argue cogently against it) - to an onlooker it just seems "if that's the best they've got to offer, it must be a logically indefensible position". This is the problem I have with Richard Dawkins, who, whilst I share many of his premises, has done the cause of humanism no end of harm in intellectual circles by failing to realise his opponents have moved on form the 19th century (shooting white elephants simply makes it look like your weapon isn't powerful enough to shoot real ones).

I spent a lot of time at uni studying "logic" - the modal kind that explodes your head but makes for great SF, the symbolic kind (whihc, God help my students) I used to teach, and the less mathematical kind - induction/deduction - Hume, Descartes, Aristotle, more inmportantly, in relation to these arguments - Popper.

On the specific point of futurology (arising from the blog), this is where Popper comes in. I made some predictions about the future of publishing. I am aware of the pitfalls of soothsaying. It would be easy for me to claim my predictions as rigorous because they are falsifiable, but that's not really good enough - future falsifiability renders things - for the present - of little other than speculative value (which is partly whta my observations were - a bit of fun). But I also wanted writers to take note of what I was saying and modify their behaviour. At that point the futurologist has to declare a methodology so that people can decide whether or not they wish to engage, act upon, or merely sit and wait. My methodology was inductive, drawing parallels between publishing and other industries. It's not foolproof. It's only "logical" inasmuch as any other statement about the future ("publishing is NOT like other industries" for example) is. But at least I set it out knowing its shortcomings, and teh levels upon which my arguments are able to enaggae and be engaged with.

A final note on this topic of (il)logic - I see you're a fan of Ben Goldacre - Bad Science should be compulsory. People should be allowed to generalise only after a certain level of exposure to it.

Thank you for getting us thinking :-)

Jane Smith said...

Dan, if you like that first sentence you should have read my original version of it--it was much stronger!

And yes, here I agree with you: I think that the Other Side of publishing frequently shoots itself in the foot by presenting poorly-constructed arguments, or by relying on fallacies to support its points.

That's not to say that mainstream publishing is without fault here: but as it currently occupies the dominant position in the marketplace, it has less to prove. For now!

I have a review of Bad Science coming up soon, by the way: I agree with you, it's an excellent book and should be required reading for all writers, I think.

Dan Holloway said...

In our house we keep the book by the bath (the place of honour). You can, of course, follow his regular informative and amusing tweets now you're a twitterer - @bengoldacre

Elizabeth K. Burton said...

One of the sales pitches subsidy presses use to hook desperate writers is to tell them an actual book sent as a submission will impress publishers and make it more likely they'll land a contract. They cite for them the legends in the business who became superstars by starting out self-publishing: Mark Twain, John Grisham, M.J. Rose.

And I'm the one left to break the bad news that what they've actually done is all but eliminated the possibility they'll place that book anywhere but where it already is.

As for factual errors in fiction, the outsourcing of much of the copyediting by major publishers means attention is paid to the mechanics of the craft while facts take a back seat. Unless one is extremely well-read in both fiction and nonfiction, the likelihood one will fail to catch a factual error increases exponentially, and most college students today just aren't that well-read.

So, they have no problem with scenes in a novel where the physics are derived from TV or films, or perpetuate a myth that slipped by a copyeditor in a bestseller and so became accepted as truth.

Jane Smith said...

Elizabeth, yes to all those things that you wrote apart from this bit here:

"They cite for them the legends in the business who became superstars by starting out self-publishing: Mark Twain, John Grisham, M.J. Rose."

I've not checked up on M. J. Rose, but John Grisham has never self-published; and Mark Twain was a success before he did so, then almost bankrupted himself by self-publication. He had to undertake a speaking tour to make a living, which he hated and which made him ill. So next time someone suggests that either of those two were self-publishing successes, you just curl your lip at them and tell them they're wrong. Ha!

As for movie-science: yes. I agree completely. It's horrible. My son is amazed by the things some of his classmates come up with in science lessons, which they believe to be fact. And as for the supply teacher who told his class all about Atlantis... actually, no. I'll dig out the emails I wrote about it and if they're not too out-of-context I might just make them a new blog post, as the whole situation was incredible.

BuffySquirrel said...

Atlantis! I love Atlantis. Two tidgy bits from Plato spawned a whole mythos!

catdownunder said...

Logic should be taught in schools - but I have a feeling politicians, advertising agents and vanity publishers would not be very happy about seeing that happen.

Captain Black said...

I'm a scientist by training and don't often fall for such nonsense as you've highlighted. I look forward to the Bad Science review...

BuffySquirrel said...

Eh, one of the first things Thatcher's government did was move the teaching of history in schools away from analysis and towards rote-learning of dates.

This wasn't an accident.

Imogen said...

Thanks for reminding us all of the need to speak and write straighforwardly, check our facts, check that the facts do support the hypothesis, and so on!

When I was twelve my grandfather lent me his copy of "Straight and Crooked Thinking" by Robert H Thouless; I later inherited it from him. There are probably a number of more recent publications on the same theme, as it was first published in the '30s, but I remeian grateful to Grandpa for it. He helped me learn to avoid being influenced by fallacies, biased sampling, sweeping statements and so on. It doesn't make it any easier dealing with those who rely on them, but at least it helps to have a chance of being aware...