Brandon Sanderson's focus on how the cost is an important part in making magic interesting reminded me of the problems I had with that. The most recent occurrence was with the magic system I was going to use in my nanowrimo novel last year. I had a magic system where people could focus their aura into various parts of their bodies to become strong/faster/better (inspired by the magic used in Hunter x Hunter for those of you that read manga or watch anime). The usage part of the system was working out fine. I felt like I had a good system were the limits on what they could do was easy for the reader to see and the visual aspect of the magic made it fun to write about. The only problem left was how to make the magic users not omnipotent compared to a non-magic user. If they could go around and enhance their own abilities the whole time it would create a strange power balance in the world.

In the story I wrote for nanowrimo I ended up using the simplest solution; they get very tired after prolonged use. That restriction turned out not to be enough. The nanowrimo story spiraled out of control as I wrote it. As soon as the main character learned how to use his own power I realized that any interesting challenge would require the antagonist, or his minions, to have the same magical power as well. This in turn created a world were too many had magical powers, while I had created the setting based upon the thought that the magical power was a rare attribute. Thus the whole world, and all the challenges that the protagonist met, started to feel very contrived.

I realized that the abundance of magic was the source of the problem, but until I read the post on Sanderson's Second Law of Magic I had not been able to pin point the exact cause. It had to be harder to use the magic.

Now that I know the cause of the problem I'm looking forward to reworking the setting, and seeing what will happen to it when I add a better cost to the magic system. Not sure what the cost is going to be, but now that I know where to look I should be able to find something fitting.

Trying to find an answer to those questions was one of the reasons I wrote the last post; to help organise my own thoughts around magic systems and thus help me create better ones for my own fantasy settings.

The origin of the structure isn't really that important in my opinion. I care more about the actual structure and how it has (and will) affect the setting. If the magic is an inherit part of the world, power from another realm, or something given to people by their Gods does not really matter.

If the system is carefully laid out with mathematical formulas I, as a reader, don't need to know that. I want enough information to predict results and options within the system, but an approximation is enough in that regard. The difference between knowing the character can create a 10" fireball and a 11.5" fireball is not that important (in most cases), knowing the character can create a fireball is the important part. For the author it might be a good idea to lay out the rules with all the small details made clear. I know that I at least have an easier time keeping track of everything, and figuring out all the possible implications of my system when I have it clearly written in front of me.

Balancing the amount in information, and figuring out how to give is a very difficult task. The reader doesn't want an info dump with all the rules, but he should learn enough to anticipate the limits and available options set by the rules. I've noticed that it is often solved with the character learning about the system alongside the reader (e.g Brandon Sanderson in the Mistborn books or J.K. Rowling's approach in the Harry Potter books), but that option is not always available. I cannot think of a book that does it any other way at the moment, without info dumping, but there has to be some out there.

You can of course avoid all of this by making magic a secondary part of the story. If it is never used to solve any conflicts it does not require the same amount of exposition.

As I was rereading the conversations my characters were having I realised that they almost always ended up with one person acting as an information dump, while the other person asked (bad) questions to keep the information coming. Often this was information that both characters should have known, so the whole sequence was just to feed the reader with information.

I solved a few of those cases by moving the information to the thoughts that the point-of-view character had, or by removing the dialogue all together. The sequences became a lot better, but it reduced the amount of dialogue a bit too much. Dialogue can tell you a lot about a character and it is a great way to convey the difference between characters, so I did not want to remove too much of it.

As I listened to the pod cast from Writing Excuses I learnt another way to make my dialogues better. I should write each dialogue as an argument between the participants. This piece of advice also fit perfectly into the general advice that each scene in a book should have some sort of conflict, and an argument is the natural way to add conflict into a dialogue heavy scene.

Rewriting the information dumping sequence to become an argument forced me to flesh out the characters, and once that was done the information that I required felt like a natural fit on the page. The cases were I needed to convey more information I knew something more fundamental was wrong. At least the times I have come across so far have been signs of me trying to force the characters to take actions based only upon what I need for the story, not upon what would be natural for the characters.

This realization has caused me to rewrite large parts of the book as the characters have become more defined and I have gotten a better idea of what information both the reader and the characters require at the various parts of the book. At least I know this trick now and can take it into consideration from the beginning when I start my next book.

The magic system has always been one of my favourite parts of a fantasy setting. Magic is one of the things that set fantasy apart from other types of fiction, and as a kid I would often loose myself in daydreams of having magical abilities. But even though I have always liked magic I have never thought too much about what sets a good system apart from a bad system.

I have always seen a well defined system of magic, with clear rules and boundaries, as a good thing in. It makes it easier for me to associate with the magical users in the story, and it gives me another tool to use when I try to figure out how the characters are going to get through the next chapter. It is also important that the magic comes with a price, both when using it and when learning how to use it. Finding a balance between that and the amount of magic you want to have available in the world is a really hard balance act. Too little and you will end up with everything done by magic. Too much and no one uses magic. Each of the two extremes might work in their own stories, but I tend to find the magic systems that fall a place in between the most exciting to read about.

A great example of a well defined system that makes the story better is the system used in Brandon Sanderson's Mistborn. The rules for the magic is well defined and explained in the beginning of the books, so when it is used to solve problems later on you are amazed by the creativity of the characters as the apply the know rules in ways you would never have imagined.

My stance on the unclear systems is a bit more vague. Some of them I like, some I don't. Until I heard those episodes from Writing Excuses, and through them got introduced to Sanderson's first law, I had not tried to articulate my preference either. And after thinking it through I agree with Sanderson's approach to magical systems. An unclear system is fine, but it can only be given a minor role in the book. If an unclear system is given too great of a role in the story, it becomes a deus ex machina and hinders the readers from investing in the characters and the story.

I think that is why I love the use of magic in Tolkien's work, even though that is a very unclear and undefined system. On the other hand you have the magical system used in the Sword of Truth series by Terry Goodkind, which I did not like. The magic in that series is unclear and often overused, so it comes off as a crutch to keep the main story going more than anything else.

One thing in particular has struck me lately, as I have been struggling with myself. How do you end a dialogue?

Writing dialogue is hard enough to begin with. It has to sound natural, while leaving out all the boring and repetitive parts that we actually say. Ending the conversation is not that hard when the characters talking walk away from each other, then the ending comes naturally.

The problem comes when the interesting part of the conversation is the beginning, and then the rest of the conversation is them doing small talk while they finish their dinner. Do you abruptly end the dialogue section with "and then talked about the weather for the rest of the dinner" once the interesting part of the conversation is over? That might work if not done too often, but I think the reader would start to feel left out if it happens all the time. Every once in a while it might be possible to end the scene and change to the next interesting event after the dialogue is done, but that feels like a brute force method as well.

After I became aware of the difficulty I was having with dialogue endings I started to take note of how other authors did it. Some kept their conversations going on forever, but with enough flair and interspersed action to keep it interesting. While impressive, that is not something I am hoping to accomplish just yet. At times something happened that forced a stop in the conversation, like the room they were in catching on fire. But most of the time the scene would come to a natural conclusion right after the dialogue ended.

So maybe that is the great skill that I have overlooked. I place the conversations I write in the wrong context. Instead of having the conversation only last half of the dinner, I should spend more time explaining everything around the talkers, so when the last line is said dinner is over and they can go and do something else.