A Study of Ice and Fire

Erik Germani

A Song of Ice and Fire has been talked to death. Between the TV adaptation (a ratings dreadnought, dragging in its wake a froth of thinkpieces), the source text (sprinkled with enigmas to entice even the most discerning of crackpots), and the book releases (rarer than Olympics), a wealth of criticism has accumulated for any fan of the series to explore.

Interested in the mysteries of the Pink Letter, fAegon, Coldhands, or Howland Reed? The tea leaves have been read, the entrails inspected: help yourself to some tinfoil. Curious about characters? Not only have their psyches been plumbed, but their maladies have been diagnosed.Trimethylaminuria, anyone? If that’s not your bag, you can read up on Westeros’s economy, climate, religions, geopolitics, genealogies, and heraldry.

But what if you want to know how the books are written? What Martin’s skills are as a writer? What exact type of lightning is in this bottle?

That last question is most interesting to me. I want to know what makes these books special to so many people, including my younger self. AGOT is the one time I read something, thought, “that was phenomenal,” and then watched it go on to become a phenomenon.

We’ll explore that phenomenon through the lens of Martin’s craft: his prose, narration, characterization, and plotting. That means most of this study is aimed at writers, but just like any feast in ASOIAF, there are a lot of courses here, and you can pick and choose what seems interesting to you.

Besides writers, I imagine there are two other potential audiences for this sort of study:

So besides providing a short description and word count for each piece, I’ve also tagged them according their appeal. If you aren’t interested in writers’ shop talk, but like literary criticism, go for the red bullet. If you like quantitative criticism, go for blue. These essays were written in no particular order, and can be read at random, but I’ve tried to put the more novel and interesting analysis towards the top, fearing that my poor lede-writing might cause the good stuff to be missed.

chosen-one-coverSome housekeeping before we get started: I’ve got a book (that’s the cover to the right), which might appeal to an ASOIAF reader. I’ve also got a Twitter account if you’d like to get in touch. This kind of analysis would have been much harder without Tower of the Hand and A Wiki of Ice and Fire, so thanks to the people behind those. I would also like to thank Dave Liepmann and everyone else behind Tufte CSS. I spent too much time working on this text, and it was nice to have the CSS already taken care of.

Charts, Tables, Graphs

Table of Contents

The Faceless Men

I tagged every line of dialogue in the series to see who talks with whom, and exactly how much. I use that conversational data to prove a very easy point: that the later books aren’t as good as the first three.
(2911 words)

The Unseen Plane Crash: Psychic Distance

Psychic distance is the most important part of the narratorial voice. I dive into Martin’s, and question just how close his point of view is.
(3913 words)

Bodies are Essence

A technical consideration of Martin’s typecasting, which bleeds into a study of Varys and neuter characters in Westeros.
(3655 words)

George R.R. Martin, The American Homer

I was all ready to do a hatchet job, but then I read some sentences that made me reconsider Martin’s prose style. I examine the effect the oral tradition has on Martin’s work.
(1693 words)

The Redemption of Jaime Lannister

An exhaustive study of Jaime’s arc in ASOS. I cover buddy comedies and character foiling.
(6198 words)

Linguistic Anachronism

I cross reference every word in ASOIAF against an etymological dictionary, to see how important diction is to creating convincing fantasy prose.
(1896 words)

The Mycah Incident

I argue that this is the most useful scene in AGOT for what it does to set up the characters.
(1511 words)

Soap Opera and the Modal Cluster

I compare ASOIAF to a soap opera, and stumble across a great way for a computer to find the plottiest parts of a novel.
(658 words)

Pfister’s Characterization

A narratologist named Manfred Pfister created a taxonomy of characterization. I use examples from AGOT to explore that, and consider the many ways we can communicate character.
(2348 words)

Inside the House of Black and White

How you make gray characters, using Jaime as an example and printing as a metaphor.
(1394 words)

Making a Memorable Character

Armed with a complete script of every characters’ dialogue, I wade in and see who makes a bigger impact than their pagetime would suggest. I offer a simple rubric for making memorable characters.
(1966 words)

Compressible Paragraphs

Of the three studies I’ve done, this is easily the strangest feature I’ve found.
(732 words)

Imprecise Sentences

They can’t all be winners.
(1045 words)

Nice Moves

A miscellany of nice moves from AGOT.
(896 words)

Clustering Characters

I use unsupervised learning to discover character types in ASOIAF. It works – with plenty of supervision.
(1028 words)

Now, As We All Know

How do we get the reader up to speed on the backstory?
(2692 words)

Zeugma and Literary DNA

While following up on a sentence structure I discovered in a prior study, I find some correspondences in Martin’s work.
(655 words)

Cultural Anachronism

A brief look at the challenge of being funny at a Ren Faire.
(651 words)

Line Reading

Martin has an ear for dialogue, and he knows how to control your ear for his dialogue.
(463 words)

Arya and Sansa

A close read of a scene between Arya and Sansa.
(2081 words)


A quick lesson on when to use a semicolon versus a colon.
(419 words)

Your Reputation Precedes You

Martin tells you about characters before you meet them, and I explore the effect that might have.
(1114 words)

Purple Prose in Green and Red

ASOIAF isn’t gritty all the time. I dug up some of the poetic sentences, and speculate on why they survived revision.
(587 words)

The Narrative Value of Dreams

We all have them, and we should probably spare our friends and family the burden of having to hear about them. In this short piece I look at how Martin uses dreams as a plot lever.
(404 words)

Riding the Waves

I don’t like character arcs, but I do like character waves. A riff on a Pynchon lecture.
(836 words)

Scene Types

A taxonomy of scenes, and how Martin moves between them.
(574 words)

The Short-Hop Flashback

One of Martin’s go-to moves, and a good way to break the “show, don’t tell” convention.
(814 words)

Odds and Ends

A collection of sidebars, half-thoughts, and quotations without much analysis.
(7101 words)