Three points make a story arc

If you've been around this writing thing for very long, surely you've heard of the "story arc." People talk about character arcs, and plot arcs, and how one should build and sustain them. I talk about them myself, because I can feel them in my work as I write - but I know of a number of people who have difficulty with the idea of the arc, in part because they say they don't know what an arc is and how to detect one.

I have a very simple definition for you. A story arc is created when something in your writing gets repeated three times. In geometry, you use three points to plot (and confirm) a line; in a story, three points create an arc. In geometry, if you only plot two points, you can't be certain you haven't made an error, so the third point confirms that you've plotted the line correctly. In a story, if you've put in two points, then people will sense an arc forming and be looking for a third point.

Here's an example. I have an unnamed character appear in a concert scene chapter 1 of my book; the protagonist notices him as a person out of place. I then have this same character appear at a concert in chapter 3; the protagonist isn't sure this is the same guy, but approaches and speaks to him. Then in chapter 10 he goes to another concert. This is "strange guy appears at concerts" arc, and I expect that when my protagonist gets to the concert, he'll be unconsciously looking for this guy - as will my readers. He won't be there, but he'll show up in another place. This very minor character therefore has an arc in my story. His first appearance sets him up, his second appearance (and the fact that more happens in it) establishes that readers should be looking for an arc, and that way when you want to put in the next piece, you know people will be looking for it.

That's a very minor arc. A major arc can have many more points contributing to it. Each of my characters in For Love, For Power has two major arcs: a love arc, and a power arc. It makes sense! What this means to my writing is that in every chapter, I need to make sure I put in something to advance the point of view character's love arc, and his power arc. If I don't contribute to each in some way, then readers will have the sense that I'm dropping balls and I am not helping them to keep track of things properly. However, because these are large-scale thematic arcs, I can contribute to them in all kinds of different ways. Tagret's love arc can involve his friends, or his mother, or his girlfriend - and each of those can be considered as its own arc, but because they all contribute to the larger thematic arc, I can take turns developing them. Tagret's power arc can involve his father, or his brother, or other politicians.

So to define arcs, we need three points of repetition. This repetition is flexible. It can be a single object. It can be a single character. Or it can be a type of behavior on the part of a character (such as risk-taking). It can be a kind of thought or worry that the character has. But so long as we can be detected as returning to whatever this thing is, we can be construed as creating an arc. Arcs can also have subordinate arcs inside them, such as when instances of Tagret seeing his girlfriend contribute to his larger thematic love-arc.

You might be asking, "Do readers really know we're doing this?"

Well, they may not know it consciously, but humans are very good at detecting patterns of repetition on a subconscious level. It's one of our language-learning skills, and we do it all the time without thinking. In fact, as writers we are usually creating arcs subconsciously rather than deliberately. However, it helps for us to put a label on them and deal with them consciously, because if we do this, it's much easier to keep them under control.

Take a look at your draft. Try to identify the repetitions. Similar situations, or characters, or behaviors, or imagery. That's where you're creating your arcs. Here are some things to look for.

1. Am I only putting in two points of the arc? If so, chances are your readers are looking for a third, and will miss it if it's not there. Either complete the arc, or break it by removing the second point.
2. Am I precisely duplicating the first point with my second point? If so, the repetition may feel strange or uncanny. In an arc, usually the second point develops on the first (if only slightly).
3. Can I put a name on the larger thematic arcs? If you can, it will really help you move your outline and your book forward. If you can't, you might be wandering off the core point of your story.
4. Am I regularly contributing to the larger arcs? There are a lot of ways of doing this, but if you can identify the larger arcs, it will become easier to contribute to them and keep all those balls in the air.
5. Do I have a lot of similar small arcs? If you're struggling with the question of larger arcs, see if you can find patterns in the smaller arcs that you're using. That can help you detect larger patterns and get a better orientation on the overall trajectory of your project.

I hope I've given you some ideas that will help you articulate with the issue of the story arc. Arcs are a very important conceptual tool, and extremely useful, so good luck with them!

Am I holding back from what the story demands?

You've seen it in your own reading. The story conflict is progressing, and you think it's going to lead you to a really incredible, horrifying, mind-blowing place... and then it never goes there. The scene cuts when you don't want it to and suddenly you're in a different point of view on the outside looking in; or the climax you were expecting turns into something else, which will necessarily be an anti-climax because of the amazing thing you thought you were going to see.

So how do you recognize it - and avoid it - when it happens in your own writing?

There are two ways to approach this: from the personal direction, and from the story direction. From the personal direction, you need to take a look at yourself and identify your own areas of sensitivity - those areas where you're most likely to hold back. For me the main areas are death, sex, and violence. Probably no big surprise. Because I'm very sensitive about those things, it can make them very effective in a story when I use them properly, because the intensity of my reaction to them will come through in the writing. However, my sensitivity means that I'm not the best judge of when I have gone too far, or when I haven't gone far enough.

So, having identified the areas that I need to look out for, I then take a look at the story and try to determine what the story demands. A lot of my sense of this develops while I am writing, and my instincts to follow the principles of the society I'm working with, or the conflict that I'm developing, will take me up to the edge of a sensitive issue - and drop me there wondering what to do next. A big scene of violence isn't something I can just write and plop into my story. It's unnecessary, and gratuitous. But if I'm digging in deep, following the characters, their psychology and their motivations in the context of the restrictions that society puts on them, I can arrive at a point where such a scene grows naturally into the story, and indeed, is demanded by the story so that the feeling of anticlimax doesn't result.

I'm at a point like this right now in For Love, For Power, which is why I'm writing about it.

My story says, "write a scene where Nekantor's gang roughs up prostitutes in a brothel that caters to the noble caste." My admittedly prudish mind says, "eek!"

This is where I go to the next suggestion: get an outside judge. If you can, find someone who is an ideal reader for the piece you're putting together, who knows the vibe of your story and can understand where it's going on its own terms (and not the ones in your head). Again if you can, try to talk it out with them and gauge what level you want to set the scene to - before you've actually attempted to write it.

In my case, I went to Janice Hardy, because she gets this book (and has from the very first partial draft). She pointed out to me a couple of things that are working in my favor: 1. Nekantor is the antagonist, and 2. he doesn't see the world the way I would see it. I think I can use those things to make sure that the moral compass of the story isn't lost when we come into this section, and do a better job of striking it just right. It's not only what happens in the scene that's important, but how the context surrounding it, and the judgments surrounding it, cause it to resonate with the whole.

Of course, I could just try writing something, and wait for the critique stage, for someone to say, "why didn't so-and-so do this?" or "that scene wasn't what I was hoping for." However, I have a terribly hard time hazarding a guess at a scene. I prefer to approach this issue earlier in the process, because my first drafts tend to lose cohesion if I don't work things like this out in advance.

I don't know about you, but I find that the stories I create grow quickly beyond what I expected when I designed them. It's in the nature of the stories. Working in science fiction and fantasy allows a writer to set up principles on an abstract or idealistic level, and then grow them into a world where they can be operationalized on the ground. So how strong are these principles, and how broadly generalized across the culture? That's a question to ask yourself in worldbuilding, as you move toward making the story happen. Once you've decided how far the principles go, however, I think you shouldn't be afraid to discover where they lead you. If you get there and feel unsure, as I often do, ask for outside judgments. When you find exactly the pitch, tone, and approach to make the scene work - and not just work, but sparkle - it will be worth it.

Do I have a story?

I have a story. This is not the same thing as saying, "I have an idea." In my particular case, I've been searching for this story for a while because numerous people had asked me to write it. "Write a sequel to Cold Words," they said. Easier said than done.

My linguistics stories are like puzzles, with lots of complicated interlocking parts, so as a result I feel I have to know where they are going or I can't even start. "Cold Words" was like this. The idea of writing a sequel had several advantages over starting from scratch:

1. I wouldn't have to design a world with aliens.
2. I wouldn't need an entirely new main character.
3. I wouldn't need an entirely new alien language and an entirely new voice.

All of this is an incredible time-saver. But as many who have tried to write sequels probably know, there are also some serious disadvantages.

1. Any subject of a major revelation in story one has to be a starting premise for story two. You need to inform readers of the "secret" (or former secret) in case they haven't read the prequel, but it can't be the point, or the story will just be covering the same ground.
2. Having a world and a character, and even quite a clear sense of chronology and what would have happened after the first story ended, does not mean you have a story.

This number 2 can effectively stop me in my tracks, because until I have a story, I won't start writing (this is not the case for all writers!). I might also remark, of course, that the lack of a story hasn't stopped many sequel-makers, at least in the case of certain movies I have watched.

What is a story? I'll sum it up this way:

Character X discovers Y and must accomplish Z before A happens or else B.

This rather closely resembles the elevator pitch for a novel. It distills story essence down to its bare bones.

In my case, I've been gathering elements that I like - Rulii and his dependency on humans, his relationship with Parker - and questions that people have asked me - about gender relations among the Aurrel, etc - and trying to fit them together. At last I've gotten somewhere. I don't want to spoil any surprises, so I'll just say that Character X is still Rulii, Y involves Parker, Z involves Parker and an Aurrel Female, and B has to do with human presence on Aurru. It's always good to ask yourself, "What's the worst thing that could happen to my character now?" and try to have it stand in that B position. That makes for a much more exciting story.

In fact, it's good to ask whether you have a story even if it's not a sequel, though sequels are probably more prone to the lack-of-story problem. Giving your character goals and stakes makes everything more exciting.

World Building is not just a Genre Issue

My husband just finished reading a book last night: Stieg Larsson's The Girl with the Dragon Tattoo. He loved it, and he thought the translation was wonderful, and he was telling me how impressed he was that they didn't try to explain everything, but actually let things like the name of the supermarket just slip in naturally to the flow of narrative.

I blinked at him a little and said, "Why would they try to explain it?" Everybody who has been around my blog for awhile knows how I feel about infodumping.

It did bring something to my attention, though - something which has been growing in my consciousness for some time now. It is this:

World Building is not just a Genre Issue.

Look at The Girl with the Dragon Tattoo. It takes place in Sweden. The author was Swedish - makes perfect sense. But it has had enormous, worldwide success. For people outside of Sweden, Sweden is another world. This book would not have taken off as it did had it not had terrific, low-key worldbuilding of the Swedish environment. Without a solid setting, you feel lost. Okay, so the author was writing his own familiar environment, but from the perspective of many of his readers, he's creating an entirely new world. And he's doing it marvelously.

If you want to have a really wonderful book, the setting can't be generic. It has to be an integral part of the story, and it has to have meaning, depth and life. This is true no matter where or when it is: historical contexts, fantasy contexts, alien worlds, or around the kitchen table. And if you can think through your worldbuilding systematically and make it really strong, then that will help the story transcend the audience that would be most familiar with the environment you're working with. Some might guess that a story that takes place in Sweden would appeal only to the Swedish - but obviously not so.

There are a lot of elements that go into making a successful story that can reach a worldwide audience. But I would argue that richness of world is a vital element on that list. And I'll conclude by saying it again:

World building is not just a genre issue.

Should I write to a market?

If you spend a lot of time visiting writers' forums, you may encounter differences of opinion on whether to "write to a market." For those who may not know, writing to a market essentially means letting the market you wish to sell to dictate how the story you write will work. This can happen either before or after you have actually drafted the story: some writers will read quite a bit of material from a particular market to get a sense of what those editors might like, and then attempt to craft a story that fits those parameters; others will have an idea, sketch it out, then pick a target market and (again on the basis of reading what has sold to that market) tune the story to fit. The fit can be stylistic, content-based, or even one of length.

The argument for writing to a market goes something like this: "Of course you should be aware of what a particular editor likes before you send something to him or her. If you don't keep those editorial tastes in mind as you write, your chances of a rejection will go way up."

The argument against writing to a market goes something like this: "Of course you should be aware of what a particular editor likes before you send something to him or her. But chances are you'll be less true to your story, or lose touch with your Muse, if you try to cater to editorial tastes too directly. Your chances of a rejection will go way up."

Both of these agree on two points: 1. It's good to be aware of what an editor likes, and 2. Your chances of a rejection are quite high.

I confess I have difficulty writing to any particular market. When a story idea jumps into my head, I have to write it the way I have to write it. Some story ideas demand a higher word count for me, and others a lower. Some story ideas call for more description, and some for less. Some call for lush voices, and some for spare ones.

You could say that I write to a market now, because I do design stories specifically for Analog magazine. On the other hand, the fact that I wrote a story that Analog wanted to buy was pure coincidence. It was a story that asked to be told, and I had a great tip from Sheila Finch on where to send it. It was only after that that I decided I should deliberately look around for stories to tell about language and culture for the purposes of sending them to Analog. Fortunately for me, I find that alien design, and language and culture stories are incredibly inspiring. The fit is natural.

If you are not going to plan to write to a market, then finding a place to get your story sold will be a little like the process of finding an advisor for a Ph.D. At that point in your studies it's no longer a question of whether you're a good enough student (feel free to substitute "writer"), it's a question of whether there's someone out there who wants to work with you. It is a question of fit between what you can do and what the writer or editor wants.

I can't say I don't envy those around me who talk about writing to specific markets, particularly when they appear to be able to do it successfully. I suppose one could look at submission guidelines for clues, or read a lot of material from the magazine, or find editor interviews online. After that, though, a lot of guessing is involved until you actually manage to make a sale.

Be careful that you don't lose your convictions in your efforts to do what an editor appears to want. Many editors have idiosyncratic, eclectic tastes - you never know what may appeal to them. It could be that if your story resembles some they've seen before, that they'll love it - or it could be that it will strike them as derivative.

I believe that there is wiggle room. If you have a great story with a dynamite core conflict, and you manage to keep its drive going all the way through, then you can play around with thematic and textural elements to give it a somewhat different flavor.

I have a friend, though, who told me something quite fascinating about a discussion he took part in about international science fiction. Several of the people involved came to the conclusion that international sf was often more exciting precisely because people weren't writing to a market, and thus were pushing their ideas further than they might otherwise.

I encourage people to look for highly original ideas, no matter how they execute them. Originality can be tricky, but one thing I've found useful is looking around in my personal experience for story ideas rather than looking for them in stories I've read. Of course, taking traditional story tropes and turning them on their heads is another favorite hobby of mine.

In the end, I think it's important to write what inspires and excites you. Because if you can't get excited about a story, it's hard to imagine how anyone else will. Yes, I do rather believe in the Muse, though I think she's less intractable and more cooperative than many others do.

I wish all of you the best in selling your work, wherever it happens to land. Because I love reading good stories.

The right narrator

I'm designing a new story. I always enjoy designing because most of it I can do freely in my head any time I want, and not have to worry about whether I can find the time to sit down at my computer. Those who know me well also know that I design by talking out loud - so I do a lot of design while bouncing ideas off various friends (you know who you are - thanks so much!!!).

I've talked before about story design, and the various steps I go through to get a linguistics/alien story going (linguistic problem, aliens, technology level, plot, language, to summarize briefly). This time, though, I'm not dealing with aliens; I'm writing a fantasy story about ancient Japan. This gives me real-time research instead of intensive world and language design, but in fact my attitude about getting close to the thinking process of the people involved remains almost the same. I'm at the last step now, which for the aliens was "language," i.e. deciding what kind of language to use to portray the alien point of view.

For this story I'm calling the step "narrator." I could always generalize and call it narrator for the other type of story too. It's really important to think through:

1. who your narrator is, i.e. how they allow you to get the best point of view on the core conflict of the story, and
2. what their voice is like

I'm not going to spend much time on voice here because I've posted about it several times before. I'm going instead to talk about #1.

Protagonist narrators are the most common in all the works I've read. These come in a number of varieties - I'm not so much talking about 1st versus 3rd person narration here but about whether the narration is delivered in the moment or retrospectively. (I'll talk about non-protagonist narrators below.)

In-the-moment narration works well if you're looking for a sense of visceral closeness to the story. The reader shares sensations, judgments and discoveries with the protagonist as the protagonist has them. This type of narration also allows your reader to experience the protagonist's particular form of myopia, including unreliability (due to insanity, bad judgment, or ignorance). It keeps you as a writer from having to withhold information deliberately and potentially irritate your reader.

Retrospective works well for other purposes. A retrospective narrator has more control over the flow of the story, and can orient the reader on the meta-level to the story as a whole. This allows the writer more freedom to play with chronology because it typically involves less requirement for the end of one section to stick to the beginning of the next. The protagonist can comment on his/her past judgments and orient the reader to a larger moral/general message, like saying what a fool he was when he was a kid. And I'm sure you've often seen retrospective narration used for foreshadowing with phrases like "little did we realize at the time..."

Of course, you can also use more than one narrator. I don't think that switching point of view should be done lightly - it should have a distinct purpose. Crime novels will sometimes dip into the antagonist's viewpoint in order to increase the sense of peril. I personally use point of view switches to show misunderstandings, sometimes within the very same conversation. Part of what I try to do with a switch of narrator is show my readers that no issue is as clear-cut as it seems to any one character in the story, and that even the most reliable characters can still be a little bit wrong. This is definitely the case with my story, The Eminence's Match (forthcoming from Panverse Publishing) - not one of the characters is entirely reliable.

I promised I'd talk about a non-protagonist narrator, and here I am: when you're thinking about who the narrator is, you should think through what the consequences of your choice are for what the story is about. The story I'm designing right now is about a young woman in Heian Japan who goes mad and enters an alternate world (trying to avoid too many spoilers as yet). She's most definitely the protagonist, but there are some difficulties with the idea of using her as a narrator. The biggest one is that she goes mad and becomes unable to tell what is real and what isn't. This makes her a pretty tough vehicle for the reader if I want to keep my reader oriented in the story. I could go for distant narration, but I'd lose some of that sense of closeness that I enjoy so much. So I've just decided to have the story narrated by one of the people who is trying to influence her life (save her vs. cast her down). One of the effects of this is that the story isn't about her "figuring out what is happening" any more - an advantage, since it's going to be tough for her to figure out anything at a certain point. The story shifts when the narrator changes, into a story about three different people trying to influence her in different directions, and which one is going to win. I need to make sure that I keep the focus on her, but this is still a much cleaner model, and I'm very excited about it.

I hope you find these thoughts give you some insight into your own story design process.

My character needs backstory!

I thought I'd share a story from my writing life. After months planning the details of a new story - human scenario, alien scenario, alien physiology, language, culture, technology level, etc. - I started to write, got 212 words in, and had to stop. The problem? It felt too shallow. I had a pretty good idea of what was going to happen, in the form of a scene-by-scene description, but I wasn't really feeling my main character, Lynn.

She had no backstory.

She had a job, a general background and knowledge base, but no individual history. And my characters need backstory - not that I ever insert it in pure backstory form in the text, but if they don't have backstory, then they always feel shallow.

Actually, I knew going in that Lynn had no backstory, but I thought I'd just figure it out as I dived in. Unfortunately, writing the opening didn't really tell me as much about her as I'd hoped. So I stepped back, thought about it, and realized two things:

1. Lynn's backstory should hand her an existing problem that could get worse or drive the conflict in some way.

2. Lynn's backstory should reflect my overall thematic issue choice for this story, namely, attitudes toward technology.

These are things that I think apply not just to my own characters, but to general strengthening of a character's role in any story. The character should fit the story. Whatever actions the character must take as the plot unfolds should not be easy, and it's more exciting when characters have something major to lose as well as gain (stakes).

So in this case, I had Lynn working as an engineer on a large-scale project (trying to avoid spoilers here) in which all information was considered highly valuable and worth protecting from spies and incidental observers. I had also established that Lynn was going to be played against the head of information security for the project. There was already an external risk to this information built into the outline... so I decided to get Lynn closer to a form of internal risk. That was when her backstory became clear to me.

Lynn has a relative who is a hacker, and for that reason has had trouble getting security clearance for her project. With her relative now in jail, she's been allowed on the project in a low level position. She's more gung-ho than most, but people are unwilling to listen to her suggestions because she's only been there five years for everyone else's ten. So she hacks the system to try to improve it herself. It gives her a secret - there's no way she can tell anyone what she's up to without endangering her place on the team, because with her relative in jail, no one would believe she's doing this for the good of the project. It also makes her highly invested in the project's success, so that when the external difficulties show up, she'll act to protect her work and that of her teammates. But she still won't be able to align herself with the head of information security, who constitutes a danger to her presence on this project she loves.

At this point I figure I have to put in the obligatory warning about beginning with backstory, or infodumping. Don't do this; go straight to the main conflict. The very fact that you know about a character's backstory will show up in that character's reactions to events.

Because I haven't yet started over with my story (and I will be starting over - bye-bye 212 words!) I can't say how I'll execute Lynn differently with her backstory. But I'm excited to work on it and find out.

Some Thoughts on Endings

There are a couple things to keep in mind about writing endings. One is that they must be prepared for properly. The other is that they can't be entirely expected.

These two things can actually work against one another, because if you've prepared throughout your manuscript for a particular ending, people will generally expect it. You then need to defeat their expectations. So if you reach the ending and experience reader protest, then there are two possible sources for a solution.

1. Change the preparation
2. Change the ending to beat the expectations

I've done #1 quite a lot: discovered that readers felt an ending I wrote was unsupported - that is to say, not given a proper foundation - in what came before. The solution to this problem is to go back and build a proper foundation in the earlier material. Set up the right convergence of characters and motives, set up the proper world knowledge, etc.

I've also done #2. When my plot gives a person a choice of two options - either the protagonist gets one thing or gets another - there can be a different kind of problem. Sometimes the reader can feel unsatisfied with either option, because they saw the choice coming a mile away and neither one answers the story problem particularly well. In that case, you can do one of two things: satisfy yourself with a sense of ambiguity, or look for a third alternative.

Naturally, any solution you choose does have to fit well with the preparation. One of the things that is most important to me is that the solution be the best one for the protagonist's particular character. An ambiguous ending requires both options to be just about equal in attractiveness from this point of view. And any third alternative has to have both a benefit and a price. If the protagonist has wanted one particular deal through the whole story, and at the end must decide to try for another outcome, there must be a big risk involved, and possibly a sacrifice. If your character gets to have it both ways too easily, then readers will feel like it was a cop-out, that all the difficult lead-in was let down at the end. And you don't want that either.

This stuff is very difficult - not only coming up with the options but deciding which one will serve the story best.

A couple of examples from my own work.

I had a big complicated solution planned for a novel which involved negotiating with multiple parties and then confronting a single heavily recalcitrant party to use the masses to sway his opinion. When I got there, it seemed foregone. So I decided to go back and set up the expectation for a bimodal (one-or-the-other) solution. The protagonists decide the only way to reach this recalcitrant person they must convince is to bring him someone he knows, through great peril, to convince him. Then readers ask, "Will the convincing person arrive or not?" They believe his arrival is the only solution, and then when he doesn't arrive, they find the planned ending much more unexpected (and hopefully persuasive - I'm still working on it).

In another story, I had the hero striving for a goal that he would either get or not. When I wrote it with him achieving the goal, all the difficulties that I'd set up for him seemed flat. Of course he was going to succeed, ho, hum. And anyway, readers weren't at all sure that the deal he got was the best one for him, because it came with a lot of strings. So that time I backed off and tried to get to the bottom of his motive. This is what my husband calls trying to find the real goal of the negotiation (in a business context): not looking at the offers that the two parties have put on the table, but looking past that to the actual desired outcome, and trying to find a way to satisfy that outcome independently of the existing offers. When I did this, I found a much more satisfying ending for the story.

Right now I'm in another quandary situation, looking at the ending of a story and trying to evaluate the building blocks, the lead-in, the characters' needs, and what kinds of options I have. It's never easy, but if I can manage to think about it clearly, I'll have a better story on my hands.

Have any of you had an experience like this? If you have, I'd love to hear.

Designing a Story

So Chicago has been great.  And busy.  And I'm still here, but I'll be flying back home on the 28th.  Being so busy having fun that I can't think straight does put a damper on my blogging, unfortunately.

Okay, so I've been designing a new story.  Making a sale tends to inspire me in that way.  I thought I'd share some thoughts.

When I do a linguistic/cultural story I tend to start with five questions (or so)

1.  What is the linguistic problem?

This is the toughest one.  What is the exact phenomenon that our linguist hero (whether it be his POV or not) is trying to pin down?  What is the punch line?  Without this, no story can work.

2.  Who are the aliens?

Here I'm talking about what kind of animal to base the aliens on.  Yes, you can design an alien from scratch, but it puts a huge processing demand on the reader.  If the nature of the alien is part of the central point of the story, then by all means go to the trouble of designing their physiology from the ground up.  A great example of aliens of this type is the story "Doctor Alien" by Rajnar Vajra, which appeared in the January/February issue of Analog.  I loved that story.  But because for me the way the aliens speak is the main issue, I don't want to send a lot of my reader's attention toward understanding the physical and physiological nature of the aliens.  Sometimes I like to select an animal that fits well with the language phenomenon I'm looking at, like wolves for status language.  Other times there isn't a really good parallel between an animal and a language phenomenon, so I can pick something else.  But after I pick an alien, I try to look at their diet and behavioral patterns so I can use that information to expand my understanding of how the aliens might live.

3.  What is the alien technology level?

This is one of those details that has to be pinned down, of course.  I like to try to make their technology real in an atypical way, by considering how the aliens make light, and what kind of objects they would keep in a home, etc.  This one has two sub-steps:  first designing the objects, and second, figuring out what they mean to the aliens.

4.  What is the plot?

Those who know me will laugh at this one, because honestly, the plot comes as number 3 or 4 in the list for me, every time.  Once I've got a sequence of events to work with, I continue tuning it throughout writing and rewriting.

5.  What is the language?

This is not the same as the language problem - it's the structure of the language, mostly the phonology and morphology.  That means the sounds and the way the sounds are put together.  I also have to know aspects of the language that relate to the language problem listed above, but this is where I have to figure out how the alien physiology meshes with the sounds they make, what words they might use in the context of the story, how the names and titles work, etc.

These are of course only the entry points.  But the nature of these stories is so complex that I can't just sit down and start something; I have to figure out this kind of stuff first.  I know I'm getting close to sitting down when I start hearing names and alien phrases in my head, and seeing a scene where two entities are talking to one another.