diff --git a/book/src/02_typology/index.md b/book/src/02_typology/index.md
new file mode 100644
index 0000000..73d5749
--- /dev/null
+++ b/book/src/02_typology/index.md
@@ -0,0 +1 @@
+## Morphosyntactic Typology
diff --git a/book/src/02_typology/typology.md b/book/src/02_typology/typology.md
new file mode 100644
index 0000000..402c3bf
--- /dev/null
+++ b/book/src/02_typology/typology.md
@@ -0,0 +1,27 @@
+### Morphological Typology
+
+L'ewa is a analytic language. Every single word has only one form and particles
+are used to modify the meaning or significance of words. There are only two word
+classes: content and particles. 
+
+#### Alignment
+
+L'ewa is a nominative-accusative language. Other particles may be introduced in
+the future to help denote the relations that exist in other alignments, but I
+don't need them yet.
+
+#### Word Classes
+
+As said before, L'ewa only has two word classes, content (or verbs) and
+particles to modify the significance or relations between content. There is also
+a hard limit of two arguments per verb, which should help avoid the problems
+that Lojban has with its inconsistent usage of the x3, x4 and x5 places.
+
+As the content words are all technically verbs, there is no real need for a
+copula. The ka/ke/ku series can also help to break out of other things that
+modify "noun-phrases" (when those things exist). There are also no nouns,
+adjectives or adverbs, because analytically combining words completely replaces
+the need for them.
+
+Nouns and verbs do not inflect for numbers. If numbers are needed they can be
+provided, otherwise the default is to assume "one or more".
diff --git a/book/src/02_typology/word_order.md b/book/src/02_typology/word_order.md
new file mode 100644
index 0000000..92c7f70
--- /dev/null
+++ b/book/src/02_typology/word_order.md
@@ -0,0 +1,57 @@
+### Word Order
+
+L'ewa is normally a Subject-Verb-Object (SVO) language like English. However,
+the word order of a sentence can be changed if it is important to specify some
+part of the sentence in particular.
+
+I haven't completely finalized the particles for this, but I'd like to use `ka` to
+denote the subject, `ke` to denote the verb and `ku` to denote the object. For
+example if the input sentence is something like:
+
+```
+/mi/ /mad.sa/ /lo/ /spa.lo/
+mi   madsa    lo   spalo
+ I   eat      an   apple
+```
+
+You could emphasize the eating with:
+
+```
+/kɛ/ /mad.sa/ /ka/ /mi/ /lo/ /spa.lo/
+[ke] madsa    ka   mi   lo   spalo
+V    eat      S    I    an   apple
+```
+
+(the `ke` is in square brackets here because it is technically not required, but
+it can make more sense to be explicit in some cases)
+
+or the apple with:
+
+```
+/ku/ /lo/ /spalo/ /kɛ/ /mad.sa/ /mi
+ku   lo   spalo   ke   madsa    mi
+O    an   apple   V    eat      I
+```
+
+L'ewa doesn't really have adjectives or adverbs in the normal indo-european
+sense, but it does have a way to analytically combine meanings together. For
+example if `qa'te` is the word for `is fast/quick/rapid in rate`, then saying
+you are quickly eating (or wolfing food down) would be something like:
+
+```
+/qaʔ.tɛ/          /mad.sa/
+qa'te             madsa
+is fast [kind of] eat
+```
+
+These are assumed to be metaphorical by default. It's not always clear what
+someone would mean by a fast kind of language (would they be referencing
+[Speedtalk][speedtalk]?)
+
+[speedtalk]: https://en.wikipedia.org/wiki/Speedtalk
+
+L'ewa doesn't always require a subject or object if it can be figured out from
+context. You can just say "rain" instead of "it's raining". By default, the
+first word in a sentence without an article is the verb. The ka/ke/ku series
+needs to be used if the word order deviates from Subject-Verb-Object (it
+functions a lot like the selma'o FA from Lojban).
diff --git a/book/src/SUMMARY.md b/book/src/SUMMARY.md
index 886a69f..d235553 100644
--- a/book/src/SUMMARY.md
+++ b/book/src/SUMMARY.md
@@ -5,3 +5,6 @@
   - [eBook Version](./intro/ebook.md)
   - [Changelog](./intro/CHANGELOG.md)
 - [Phonology and Writing](./01_phonology/index.md)
+- [Morphosyntactic Typology](./02_typology/index.md)
+  - [Word Order](./02_typology/word_order.md)
+  - [Morphological Typology](./02_typology/typology.md)
diff --git a/book/src/build.sh b/book/src/build.sh
index 8755a5e..2cf4f58 100755
--- a/book/src/build.sh
+++ b/book/src/build.sh
@@ -8,6 +8,10 @@ intro/index.md
   intro/CHANGELOG.md
 
 01_phonology/index.md
+
+02_typology/index.md
+  02_typology/word_order.md
+  02_typology/typology.md
 '
 
 set -x
diff --git a/nix/sources.json b/nix/sources.json
index 685fbf8..5954f22 100644
--- a/nix/sources.json
+++ b/nix/sources.json
@@ -1,4 +1,16 @@
 {
+    "easy-dhall-nix": {
+        "branch": "master",
+        "description": "Derivations for easily downloading Dhall binaries and putting them to use.",
+        "homepage": "",
+        "owner": "justinwoo",
+        "repo": "easy-dhall-nix",
+        "rev": "27edaa0c769b9c876f11c50c9ab542061d07def5",
+        "sha256": "0w2ckl8fifnaq12ycqqa9ldqjsys3jbblbg7f1vkqa272y89zl08",
+        "type": "tarball",
+        "url": "https://github.com/justinwoo/easy-dhall-nix/archive/27edaa0c769b9c876f11c50c9ab542061d07def5.tar.gz",
+        "url_template": "https://github.com/<owner>/<repo>/archive/<rev>.tar.gz"
+    },
     "niv": {
         "branch": "master",
         "description": "Easy dependency management for Nix projects",