View Full Version : Deus Ex Machina, Occam's Razor, Chekhov's Gun

Alexei P
2009-01-08, 12:47 AM
...or The Thread where Alexei P comes out like a total dork. Whichever you like. :smallsmile:

Being a semi-lurker here on the Erfworld forum, I've noticed how some terms, namedly Occam's Razor, Chekhov's Gun and Deus Ex Machina, have had a real popularity boost in discussing the current storyline. People would say "Occam's razor dictates..." or "that kind of Deus ex Machina ending would totally suck..." etc. People seem to really enjoy these expressions.

So, rather than derail several threads by quoting people and throwing out comments like "Occam's Razor dictates no such thing, you #$#@!", I thought maybe we could have a thread to discuss these terms and how they are pertinent to Erfworld.

Here's what I know (or think I know) about these terms:

Ockham's Razor

"Elements should not be multiplied without necessity" - William of Ockham

"That is better and more valuable which requires fewer, other circumstances being equal" - Robert Grosseteste.

"Everything should be made as simple as possible, but not simpler" - Albert Einstein.

"K.I.S.S" ("Keep It Simple, Stupid") - popular phrase

Ockham's Razor (also commonly spelled "Occam's Razor") is a principle found in theology, philosophy, physics, biology, medecine, design, engeneering, and, of late, webcomic discussion. Taken in a nutshell, the argument states: "The most simple explanation is probably the best".

The principle itself, though not its modern phrasing, is attributed to the 14th century theologist William of Ockham, who made ample use of it in his writings. It should be noted that at the time of the maxim's origin, it was generally believed that all elements in nature should have a simple, straightforward explanation. This view is no longer universally held today. One of the most famous rebuttals to the Occam's Razor argument is attributed to Walter of Chatton: "If three things are not enough to verify an affirmative proposition about things, a fourth must be added, and so on". Many theorists have recognized that Occam's Razor doesn't actually appeal to logic, but to the human bias to look for the simplest possible answers in everything.

In discussing ongoing series or webcomics like Erfworld, the Occam's Razor principle gets used a lot to dispute far-fetched conjectures about story background or predictions of future events. While it often looks convincing enough, it should be remembered that Occam's Razor isn't enough to disprove a theory by itself. It merely states that, of several theories, all of which equally account for existing data, the simplest and most straightforward is most likely to be right.

For example, one poster might start a theory about Vinny having previously been Transylvito's Chief Warlord who lost his job to Ceasar Borgata and has been harboring thoughts of revenge ever since. Another poster might refute the idea, stating that, while the theory cannot be disproved with actual evidence, it is probably too "out of the blue", and invoke Occam's Razor to affirm that Vinnie and Ceasar are nothing more than what they seem: a pair of colleagues who get along spiffily. But then, a third poster might enter the discussion, pointing out Page 265, where, in panel 6, Vinny is seen looking at Ceasar hatefully while saying "I really, really hate that bastard". At this point, Occam's Razor no longer applies, since the simplest theory fails to cover all facts - and the more complex explanation seems more likely. However, at this point a fourth poster comes along, and proposes a theory that Vinny might be resenting Ceasar for stealing his one true love, Janis the hippiemancer, and supplies a page where Vinny is seen throwing a longing, sad glance in Janis's direction. Then a fifth poster comes in and proposes that the simplest solution is that Vinny was then simply feeling sad from an argument he had with Don King the night before. Then all five posters begin arguing that their theory is the simplest given all the facts, and sparks start a-flying.

Occam's Razor is the perfect card to play when proposing a theory that adequately covers all the known facts, but is less convoluted than the theory you're arguing against. However, the argument cannot, by itself, prove the correctness of one interpretation over another. That is, even though you might appear to "win" an argument thanks to Occam's Razor, the other party is still entitled to her own interpretation, however convoluted it might seem to the rest of the audience.

Occam's Razor is actually a very imperfect model when it comes to storytelling, since the author is free to make things as simple or as convoluted as she likes. One notable instance of a story where Occam's Razor has been made to stand on its head is P.G. Wodehouse's Death at the Excelsior, where the "simplest explanation" turned out to be that the murder was perpetrated using a specially trained cat whose claws had been laced with cobra poison.

Chekhov's Gun

"Dearest Aleksandr Semionovitch! I have promptly received and read your play. It is superbly written, but the architecture is a mess. Completely un-stageable. Judge for yourself: Dasha's first monologue is completely unnecessary, it sticks out like a sore thumb. It would have been useful had this monologue, very promising from the audience' point of view, had any actual bearing towards the plot or effects of the play. One shouldn't put a loaded gun on the stage, if no one intends to fire it. One shouldn't make unsubstantiated promises."

- From one of Anton Chekhov's private letters.

Chekhov's Gun is the name of a storytelling principle, coined after Anton Chekhov, a highly prominent 19th century russian writer and playwright.

The term "Chekhov's Gun" is used to describe a literary device by which an object which first appears to have little significance to the plot turns out to be critical at a later time in the story. Alternatively, the phrase is used to describe a literary dogma that a story should not include any unnecessary elements. It is often inferred that doing the opposite (including an element that promises to be important but actually isn't) should be considered clumsy storytelling.

Chekhov's Gun has proven to be a tried-and-true method of building a story and most readers today expect the rule to be followed. It should be noted, however, that Chekhov never meant the principle to be anything more than a helpful guideline. Indeed, many great writers, old and new, have been known to break the Chekhov's Gun rule left and right, even Chekhov himself. Especially inventive authors have been known to twist the principle in order to surprise the reader. For instance, the gun might turn out to be fake or malfunction when used.

A point to keep in mind when bringing up the Chekhov's Gun principle in a discussion is that, while the principle states the loaded gun has to fire, it doesn't dictate how and when. The loaded gun introduced in Chapter 1 of the story might fire in Chapter 3, or in Chapter 9, or even in chapter 14 of tome 3. Assuming it is fired at all.

Suppose for example that, in a future Erfworld arc, the characters hear rumors of a fifth, previously undiscovered Arkentool known as the Arkenslinkie. At once, Parson, Stanley, Charlie, Don King and all other important characters start trying to get their hands on it. Everyone on the forum gets busy discussing what the Arkenslinkie's powers might be and whom it will attune to when found. Eventually, thanks to brilliant strategic planning, Parson gets a hold of it first, and then... the Arkenslinkie doesn't attune to anyone. Not to Parson, not to anyone else. Parson eventually just ends up re-selling it and it disappears from the story. Everyone on the forums acts disappointed and people start accusing the authors of building unneeded expectations and breaking the Chekhov's Gun rule. But then, two months later, it is revealed that the Arkenslinkie is actually a special, cursed Arkentool that, instead of attuning to a particular wielder, starts corrupting, and eventually destroying any side who owns it, in a bid to unmake the work of the Titans and destroy Erfworld. In fact, once it has started its work, the only thing that can stop it is the combined power of the four other Arkentools. So when the side Parson sold the artifact to suddenly starts popping strange, scary units and launches a war on Parson's capital, he finds he has to rally Charlie and all other Arkentool-wielders to his defense if he hopes to make it. Chekhov's Gun has fired - but not in a way anyone in the audience expected.

I apologize for the clichéd story concept used in my example.

Deus Ex Machina

"Clearly therefore the «denouement» of each play should also be the result of the plot itself and not produced mechanically" - Aristotle

"Don't let God interfere" - Horace

Deus Ex Machina (literally "Crane-lifted god" or "God from a machine") is a literary term that, today, can be used to describe any plot device that resolves a story in a highly improbable and unsatisfying way.

The use of Deus Ex Machina to resolve a plot goes back to the time of classic Greek tragedy - and was widely criticized even back then. Originally, the term was reserved for a sudden and unexpected intervention from a divinity as means to resolve the plot of a tragedy. With time, the term came to signify any unrealistic, unlikely or unexpected ending that breaks the audience's suspension of disbelief. In fact, that last appears to be the main criterion. If, for instance, the writer manages to have a god appear and solve the story, but in a way that suspension of disbelief is kept (maybe because the story happens in a universe where gods pop about all the time, or because the god promised early on he intended to do it), it is no longer considered Deus Ex Machina.

The popular dislike for Deus Ex Machina endings comes from the human expectation that events in a story should always be "predictable" or "expected". In real life, the unexpected happens all the time - people sometimes win the lottery, terminally-ill patients sometimes recover, and huge inheritances sometimes drop out the blue. Still, when such a thing saves a hero's butt in a fictional story, it leaves most audiences with a strange, unsatisfying feeling.

Even though today's definition of a Deus Ex Machina event is very broad, it still pre-supposes two important criteria: it must come about mostly or totally unexpected, and it must be instrumental in resolving the main conflict of the story.

Suppose that at the end of Battle for Gobwin Knob, the Titans of Ark suddenly appear and chase the Coalition away. That would be Deus Ex Machina, completely by-the-book. Suppose that Parson manages to win thanks to a previously-unmentioned rule introduced in Klog #38 that all Archons become helpless if you feed them strawberry cake. That would also be Deus Ex Machina. However. Suppose that, early in the story the Titans appear to Parson in a dream and say: "Hey, dood, if you need us, just give us a call, 'K?" Then their interference is no longer Deus Ex Machina because it was foreseen. Likewise, suppose that rendering all of Charlie's Archons helpless with indigestion helps Parson only insignificantly, because Charlie didn't intend to attack on that turn anyway. In that case, the event, however improbable, stops being Deus Ex Machina, because it is not instrumental in resolving the plot.

* * *

So, uh... I suppose I've rambled away a little. Anyway, to sum this up, and maybe get at some sort of conclusion from this overly-long post, I've found this phrase from the savvy and intelligent Monte Cook (http://www.montecook.com/montejournal) that I think sums up what I've been getting at by all this:

Every rule for writing has the hidden caveat that if you know what you're doing, you can break the rule and it's not only OK, it's great. It's hard to think of a significant rule of any kind that some great writer hasn't broken to wondrous effect at some point. Great writers use incomplete sentences, they tell instead of show, they use cliches, they do all the things they're not supposed to, but they do these things deliberately and carefully. That's why no one should ever develop real rules that writers have to follow, because it would kill writing dead.

That quote really fits Erfworld, I think.

Thank you and if you've read it all... boy I'm impressed.

2009-01-08, 12:55 AM
I am in turn more than impressed. And I admire your optimism.

As I've tired of mentioning, Occam's Razor is at most a rule of thumb; I appreciate its application in science, parsimony, seeking the elegant. But it must bow to the scientific method.

dr pepper
2009-01-08, 01:43 AM
Another method of analysis, which gets used here sometimes, but which i haven't seen explicitly invoked, is Holmes's Dictum. I could go into a long exposition about that one, but to sum up: 1) you can't always be sure what's impossible, and 2) even when you think you can, not everyone will agree with you.

2009-01-08, 07:34 AM
Wow, this is a very thoughtfull and thorough description of those three mechanics. Good work.

The important thing to note, in my opinion, is that they should not be seen as unrelenting and unbendable rules.

Occams Razor only points to the fact that when you find a dead body with a knife in stuck in the chest it is reasonable to believe that the knife was the cause of death, it does not state that by no means could this be wrong.

Checovs gun states that it is not helpfull for a story to include elements that does not further the story, it does not state that any element should be used in the traditional or expected way.

I think what i am getting at is the same thing as Alexei P ends the opening post by saying: Just because the gun is not fired but rather used as a hammer in the time of need that is not bad storytelling. The same goes for the body with the knife sticking out of its chest which is discovered to have been poisoned. It does not violate any hard and fast rule of storytelling to twist or ignore these tendencies but in general the tendencies does hold some kind of value.

2009-01-08, 09:49 AM
You really took your time here, impressive analysis of the "rules" of writing.
I completely agree with your statements, especially with Ockham's razor, people are misusing that term everywhere, not just on this forum.

2009-01-08, 08:08 PM
Because it sounds so cool :smallamused:

Well, like the other two, really.

2009-01-09, 11:28 AM
One notable instance of a story where Occam's Razor has been made to stand on its head is P.G. Wodehouse's Death at the Excelsior, where the "simplest explanation" turned out to be that the murder was perpetrated using a specially trained cat whose claws had been laced with cobra poison.
Oh, thanks a lot Spoily McSpoilerton! :smallbiggrin:

2009-01-10, 03:02 AM
Your definition of Occam's Razor is what most people use; however, that's not exactly what it means.

Occam's Razor on Wikipedia (http://en.wikipedia.org/wiki/Occam%27s_Razor)

"This is often paraphrased as "All other things being equal, the simplest solution is the best." In other words, when multiple competing theories are equal in other respects, the principle recommends selecting the theory that introduces the fewest assumptions and postulates the fewest entities. It is in this sense that Occam's razor is usually understood. This is, however, incorrect. Occam's razor is not concerned with the simplicity or complexity of a good explanation as such; it only demands that the explanation be free of elements that have nothing to do with the phenomenon (and the explanation)."

I've been using the colloquially understood version, because people understand that. Taking one of the examples from the Wiki, you're a cop and come across a body with a knife in the back. Proceeding on the belief that the person drowned is not the best course of action, because you have no evidence of drowning. Look for witnesses that saw a guy with a knife. When you get a report back from forensics that says the guy drowned, and the knife was inserted after tdeath, then you concern yourself with drowning. 99% of the time, your hunt for the knife wielder will prove to have been the best course of action, because it is vital that you get looking for the killer fast. Waiting hours or days for forensics Cause of Death lets the knifer flee the country. (I shoudln't have to post the 48 hours theory, and unlike forensics shows, it takes days or weeks to get test results, not minutes.) You don't have the benefit of time, so you must start without proof for your beliefs, because the alternative (an escaped murderer) is not acceptable.

Anyway, point is, if you're going to proceed without proof, Occam's Razor, the real version, means not adding things that are not supported by current known facts, if possible. A solution that is mind-bogglingly complex, but fits without any added elements is better than introducing a single element that has no proof that might simplify the entire picture. So, the standard definition for Occam's Razor isn't quite right, but it does fit with Occam far more often than it does not.

Just note that I don't confuse issues by using the correct version. I use the same one most others use, or readers get confused.

2009-01-10, 04:28 PM
Interestingly enough, none of the above concepts should hold any predicitive power at all regarding a narrative in full swing. But they always seem to be used to make predictions, shoot down predictions, or complain about when a favorite prediction doesn't come to pass.

Deus Ex Machina- Until the narrative has ended or been resolved, it can't be accused of having a Deus Ex Machina. (By the way, it seems really odd to complain about unknown mechanics or physics getting in Parson's way as a Deus Ex Machina when that's a central element to the plot- Parson figuring out Erfworld.)

Occam's Razor- You should only be introducing simplicity into an argument when: 1) you believe you have all the relevant facts, or 2) You need to make an action or decision and believe you can't gain any more facts. In the middle of a narrative, you are almost always assured that neither 1) nor 2) will be satisfied!

Chekov's Gun- This has no predictive power at all. You can simply note that a story element "Should be" occuring again in the future. That's it.

2009-01-11, 08:04 PM
Resolving a situation by means at the time unknown to the readers is still risky. Parson might not have a clue about something, but that does not necessarily mean readers can't be aware of it. Continuous explaining of surprising turns of events only after said events occured (and constantly putting readers in a situation where - from their point of view - things didn't make any sense) gets irritating pretty fast. My guess is that from now on these situations will be much rarer, since Parson (and us) are about to get a look at more rules and possibilities.

2009-01-14, 08:03 PM
good job! *clap clap clap*

that was really well thought out.