I am a great lover of words, while I can seldom pick a favourite film or book or song I can tell you what my favourite word is. Currently it's tmesis and has been for a few years, not I grant you an especially useful word, but I like that it begins with 'tm' which not a lot of words do and I like that it's a word about words. When I was about fourteen or so my favourite word was elope, which I discovered by flicking through a dictionary looking for interesting words (please tell me I'm not the only person to have done that) and for some of the time in between my favourite word was serendipity, which finally brings me to today's topic because although serendipity has fallen out of my personal favour as a word, the concept it represents is still highly important to me.
So then, serendipity in research. I'm a great believer in research as a way of solving plot problems or brainstorming whole new sub-plots (or even the main plot, depending at what stage of the planning process I get stuck and go and do some research), sometimes I research something obviously connected to my plot issue, if the problem I'm having is that I don't know enough about the subject the plot relates to, for example when I wanted to know about the kind of archaeological evidence you could get from bones I read a bunch of academic papers on the subject (having to learn a whole new vocabulary in the process incidentally, which was tough going but interesting and now I know my distal from my proximal and my anterior from my posterior and the technobabble in shows like Bones actually makes sense to me). As useful as that kind of directed research undoubtedly is, it's not serendipitous research. Serendipitous research, as I think of it, comes in two forms.
Type one is entirely accidental, a side-effect of regular research whereby in looking up something you want to know now you discover something that will be useful later on for something else, I had an example of this happen to me yesterday, I was looking into the possibility of using obsidian as a building material (which I'm still unclear on, but some monks made a temple out of beer bottles and obsidian is a kind of glass, so maybe?) and discovered something called obsidian hydration dating, which is another archaeological thing that is basically about dating the age of obsidian tools and stuff by how much the surface has reacted to water (kind of the volcanic glass equivalent of going rusty I guess) and my plot point radar went bing and I ended up with half a page of notes of the 'when you get as far as editing that bit' variety. This kind of serendipitous research isn't something you can force or predict, although it helps to keep your eyes open for the kind of unrelated but useful stuff that might be floating past at any moment.
Type two is about going looking for serendipity (which is kind of against the spirit of the concept, but whatever), it's what I also call 'research at random' and it works like this, I'm stuck with something plot related, I don't know what should be happening at a certain point, or I have to get the plot from point A to point C but have no clue what the the intervening plot point B should be, or I have to fill in the details of the question mark question mark question mark bit that comes before profit in the villains' plan, something like that, and for whatever reason I don't think direct research is going to help (maybe I've already done some and it didn't, or possibly I'm aware that my problem is more creative than factual) what I do then is research something random. Picking something random to research is harder than it sounds, the temptation is to pick something that obviously could be useful, but that's not the point of the exercise, so the trick is to research something interesting, I pull something non fiction out of my to be read pile (which is full of interesting but of no immediate use stuff) or I go on a wiki-walk until I find something that looks cool and intriguing or I just go and find out what someone else is researching right now and see if that's of any use. I know this sounds like a complete waste of time (and sometimes it is) but it's amazing how often this approach actually works and if not, well that's a whole new bunch of interesting factoids that just might help out with the next problem down the line.