Lexurgy SC Cheat Sheet

This is a quick reference for the syntax available in Lexurgy SC. Each of these examples is runnable: you should be able to put the contents of the Example column into the Sound Changes box and the input words from the Effect column (to the left of the arrows) into the Input Words box, and get the results shown in the Effect column.

For a gentler introduction, see the Tutorial.

Basics

Structure

Meaning

Example

Effect

Comment

Ignored by Lexurgy. Use them to remind yourself what a rule is for.

 
# This is ignored

Simple change

Change sounds to the left of the arrow into sounds to the right of the arrow.

 
rule:
  a => i

bad    => bid
banana => binini

Condition

Change sounds only when certain other sounds are nearby.

 
rule:
  a => i / b _ d

bad       => bid
bananabad => bananabid

Exception

Change sounds except when certain other sounds are nearby.

 
rule:
  a => i // b _ d

bad       => bad
bananabad => bininibad

Condition and exception

You can use conditions and exceptions together to create complex restrictions on changes.

 
rule:
  a => i / b _ // _ d

banana => binana
badaba => badabi

Word boundary (start)

Change sounds only at the beginning of a word.

 
rule:
  a => i / $ _

abanana => ibanana
abada   => ibada

Word boundary (end)

Change sounds only at the end of a word.

 
rule:
  a => i / _ $

abanana => abanani
abada   => abadi

Insertion

Add sounds between certain other sounds.

 
rule:
  * => i / b _ d

bdana  => bidana
ababda => ababida

Deletion

Delete sounds.

 
rule:
  a => *

banina => bnin
badai  => bdi

Alternatives

Change each of the sounds on the left into the sound on the right.

 
rule:
  {b, d} => x

banana => xanana
abdab  => axxax

Alternatives (corresponding)

Change each of the sounds on the left into the corresponding sound on the right.

 
rule:
  {b, d} => {x, j}

banana => xanana
abdab  => axjax

Alternative environments

Apply the change when nearby sounds match any of the conditions.

 
rule:
  a => i / {b _ d, n _ n}

banana => banina
badaba => bidaba

Sound class

Define a list of sounds that tend to behave similarly.

 
Class stop {b, d}
rule:
  @stop => x

banana => xanana
abdab  => axxax

Sound class (corresponding)

Change one sound class to another; each sound in the left class turns into the corresponding sound in the right class.

 
Class stop {b, d}
Class fricative {v, z}
rule:
  @stop => @fricative

banana => vanana
abdab  => avzav

Wildcard

Matches any single sound.

 
rule:
  [] => x

bad       => xxx
bananabad => xxxxxxxxx

Digraph

Define a multi-character symbol that should be treated as one sound.

 
Symbol ts, dz
rule:
  {t, d} => l

badadz => baladz
tsatat => tsalal

Escape

Use a backslash to treat a character as a sound rather than part of the syntax.

 
rule:
  \* => \@

ba*a*a => ba@a@a

Combining Elements

Structure

Meaning

Example

Effect

Sequence

Change several consecutive sounds at once.

 
rule:
  {b, d} a => * i

banana => inana
abdab  => abib

Optional element

The rule applies whether the optional element is present or not.

 
rule:
  a => i / b d? _

banana => binana
abdab  => abdib

Grouping

Use parentheses to group elements together, so that other syntax applies to the group as a whole.

 
rule:
  a => i / a (b d)? _

aab   => aib
abab  => abab
abdab => abdib

Repeated element (any number)

Matches any number of copies of the specified element.

 
rule:
  b => v / _ (an)* a $

ba       => va
bana     => vana
banana   => vanana
bananana => vananana

Repeated element (one or more)

Matches one or more copies of the specified element.

 
rule:
  b => v / _ (an)+ a $

ba       => ba
bana     => vana
banana   => vanana
bananana => vananana

Repeated element (exact count)

Matches a specific exact number of copies of the specified element.

 
rule:
  b => v / _ (an)*2 a $

ba       => ba
bana     => bana
banana   => vanana
bananana => bananana

Repeated element (range)

Matches any number of copies of the specified element, within a given range.

 
rule:
  b => v / _ (an)*(1-2) a $

ba       => ba
bana     => vana
banana   => vanana
bananana => bananana

Repeated element (upper bound)

Matches any number of copies of the specified element, up to the given limit.

 
rule:
  b => v / _ (an)*(-2) a $

ba       => va
bana     => vana
banana   => vanana
bananana => bananana

Repeated element (lower bound)

Matches at least this many copies of the specified element.

 
rule:
  b => v / _ (an)*(2-) a $

ba       => ba
bana     => bana
banana   => vanana
bananana => vananana

Capture

Remember the sounds that an element matched and refer to them later.

 
rule:
  {b, d}$1 => * / _ $1

abbanna => abanna
abdadda => abdada

Negation

Matches anything but the specified element.

 
rule:
  !a => x

banana => xaxaxa
abdab  => axxax

Intersection

Sounds must match both criteria at the same time.

 
Class stop {t, d, k, g}
Class alveolar {t, d, s, z}
rule:
  @stop&@alveolar => r / a _ a

bataka => baraka
tasada => tasara

Negated Intersection

Sounds must match the first element but not the second element.

 
Class stop {t, d, k, g}
Class alveolar {t, d, s, z}
rule:
  @stop&!@alveolar => r / a _ a

bataka => batara
tagasa => tarasa

Dummy rule

Does nothing. Use this if you need a rule for other reasons but don’t want it to change anything.

 
rule:
  unchanged

banana => banana
abdab  => abdab

Multiple words

Spaces between words are treated as boundaries: $ matches them, and they break adjacency

 
rule:
  t => * / _ $
  t => s / _ a

ratatat  => rasasa
rat atat => ra asa

Between words

Matches the space between words

 
rule:
  $$ => * / _ a
  t => d / n $$ _

rat atat  => ratatat
ran tanta => ran danta

Reusable element

Define a complex element that can be referenced by name

 
Element batman (na)+
rule:
  a => i / _ @batman b
  b => v / $ @batman _

bananabad => bininabad
nananab => nininav

Features

The examples in this section all require the following declarations:

Feature low, high, front, back
Feature voicing(unvoiced, voiced)
Feature place(labial, alveolar, velar)
Feature manner(stop, fricative, nasal)
Symbol a [+low -high -front -back]
Symbol e [-low -high +front -back]
Symbol i [-low +high +front -back]
Symbol o [-low -high -front +back]
Symbol u [-low +high -front +back]
Symbol p [unvoiced labial stop]
Symbol b [voiced labial stop]
Symbol t [unvoiced alveolar stop]
Symbol d [voiced alveolar stop]
Symbol k [unvoiced velar stop]
Symbol g [voiced velar stop]
Symbol f [unvoiced labial fricative]
Symbol v [voiced labial fricative]
Symbol s [unvoiced alveolar fricative]
Symbol z [voiced alveolar fricative]
Symbol m [labial nasal]
Symbol n [alveolar nasal]

Then put any additional declarations in the appropriate sections. Remember that all the feature declarations must come first, then all the symbol and diacritic declarations, then the rules.

Structure

Meaning

Example

Effect

Binary feature

These features can either be plus (+) or minus (-).

 
rule:
  [-low -high] => [+high] / _ $

benene => beneni
bonono => bononu

Multivalent feature

These features can have multiple values, each with a distinct name.

 
rule:
  [labial] => [alveolar]

bememe => denene
pofofo => tososo

Absent value

If a feature isn’t mentioned in a symbol declaration, it’s automatically absent (*).

 
rule:
  [*place] => i

benene => binini
dofofo => dififi

Univalent feature

These features are minus by default.

 
Feature +lateral
Symbol l [alveolar +lateral]
rule:
  [alveolar -lateral] => x

benene => bexexe
dololo => xololo

Feature variable

Copy a feature value from one sound to another.

 
rule:
  [stop] => [$voicing] / _ [stop $voicing]

abtadka => aptatka
akdak   => agdak

Negated value

Match only sounds that don’t have a specific feature value.

 
rule:
  [!labial] => i

benene => biiiii
dofofo => iififi

Diacritic

Add features to a sound by putting a modifier character after it.

 
Feature +ejective
Diacritic ' [+ejective]
rule:
  [unvoiced stop]$1 $1 => [+ejective] *

appakka => ap'ak'a
attakta => at'akta

Diacritic (before)

This diacritic is written before the sound it modifies.

 
Feature +ejective
Diacritic ' (before) [+ejective]
rule:
  [unvoiced stop]$1 $1 => [+ejective] *

appakka => a'pa'ka
attakta => a'takta

Diacritic (first)

This diacritic is written after the first character of the sound it modifies.

 
Feature +hightone
Diacritic ' (first) [+hightone]
Symbol ai [+low +high +front -back]
rule:
  [+low] => [+hightone] / [unvoiced stop] _

pataida => pa'ta'ida
badaita => badaita'

Floating diacritic

Sounds and classes without the diacritic match sounds with it.

 
Feature +hightone
Diacritic ' (floating) [+hightone]
rule:
  {u, e} => {o, i}

be'ne'ne' => bi'ni'ni'
bununu    => bonono

Exact match

This sound must match exactly, including floating diacritics.

 
Feature +hightone
Diacritic ' (floating) [+hightone]
rule:
  {u!, e!} => {o, i}

be'ne'ne' => be'ne'ne'
bununu    => bonono

Inexact capture

Match a captured sound, even if it has different floating diacritics

 
Feature +hightone
Diacritic ' (floating) [+hightone]
rule:
  {u, e}$1 => {o, i} / ~$1 _

be'eneu' => be'ineu'
buunu'e' => buonu'e'

Named absent value

Use the specified name to refer to the absent value instead of the * syntax.

 
Feature tone(*lowtone, hightone)
Diacritic ' [hightone]
rule:
  [-low -high lowtone] => [+high]

bene'ne  => bine'ni
bo'nono' => bo'nuno'

Blocks

Structure

Meaning

Example

Effect

Simultaneous expression

Apply several changes, finding all possible application sites simultaneously.

 
rule:
  a => i / b _ d
  b => v / _ a

bad       => vid
banana    => vanana
bananabad => vananavid

Simultaneous expression (precedence)

Earlier expressions take precedence over later ones if they conflict.

 
rule:
  a => i / b _ d
  a => u

bad       => bid
banana    => bununu
bananabad => bununubid

Sequential block

Apply several changes, one after the other.

 
rule:
  a => i / b _ d
  Then: b => v / _ a

bad       => bid
banana    => vanana
bananabad => vananabid

Hierarchical block

Apply later expressions only if earlier ones fail to change anything.

 
rule:
  a => i / b _ d
  Else: b => v / _ a

bad       => bid
banana    => vanana
bananabad => bananabid

Propagating rule

Apply the rule repeatedly until the word stops changing.

 
rule propagate:
  dd => xx
  {cx, xc} => xx
  {bx, xb} => xx

abcddcba => axxxxxxa

Left-to-right rule

Apply the rule once starting at each character, from the beginning of the word to the end.

 
rule ltr:
  dd => xx
  {cx, xc} => xx
  {bx, xb} => xx

abcddcba => abcxxxxa

Right-to-left rule

Apply the rule once starting at each character, from the end of the word to the beginning.

 
rule ltr:
  dd => xx
  {cx, xc} => xx
  {bx, xb} => xx

abcddcba => axxxxcba

Filter

The rule pretends that only sounds that match the filter exist. Adjacency passes through non-matching sounds.

 
Class vowel {a, e, i}
rule @vowel:
  a => e / _ i

aitati  => eiteti
annanni => annenni

Deferred rule

The rule doesn’t apply when declared, but can be applied later as part of other rules

 
x defer:
  {dd, bx, xb} => xx
rule:
  :x
  Then:
  {cx, xc} => xx
  Then:
  :x

abcddcba => axxxxxxa

Cleanup rule

The rule applies once when declared, then again after every subsequent rule

 
x cleanup:
  {dd, bx, xb} => xx
rule:
  {cx, xc} => xx

abcddcba => axxxxxxa

Cleanup off

Turn off an active cleanup rule for subsequent rules

 
x cleanup:
  {dd, bx, xb} => xx
x:
  off
rule:
  {cx, xc} => xx

abcddcba => abxxxxba

Deromanizer

Convert the romanization system into phonetic notation before applying any rules.

 
deromanizer:
  kh => x

khakhi => xaxi

Literal deromanizer

Ignore all declarations until the first Then:. Use this if the romanization system conflicts with the declarations.

 
Feature +ejective
Diacritic ' [+ejective]
deromanizer literal:
  ' => h
rule:
  k => k'

'a'i   => hahi
kaki   => k'ak'i
k'ak'i => k'hak'hi

Romanizer

Convert phonetic notation into the romanization system after applying all the rules.

 
romanizer:
  x => kh

xaxi => khakhi

Literal romanizer

Ignore all declarations after the last Then:. Use this if the romanization system conflicts with the declarations.

 
Feature +ejective
Diacritic ' [+ejective]
rule:
  k' => k
romanizer literal:
  h => '

hahi   => 'a'i
k'ak'i => kaki
khakhi => k'ak'i

Intermediate romanizer

Preserve an intermediate stage of the language.

 
rule-1:
  a => i
romanizer-a:
  unchanged
rule-2:
  i => u

hahi => hihi => huhu

Syllables

Structure

Meaning

Example

Effect

Explicit syllables

Syllable breaks (.) don’t break adjacency, but you have to include them in the input words.

 
Syllables:
  explicit
rule:
  a => i / _ n

banana   => binina
ba.na.na => bi.ni.na

Syllable boundary

Matches syllable breaks and word boundaries.

 
Syllables:
  explicit
rule:
  a => i / _ .

banana   => banani
ba.na.na => bi.ni.ni
ban.ta   => ban.ti

Syllable element

Matches an entire syllable.

 
Syllables:
  explicit
rule:
  <syl> => x

banana   => x
ba.na.na => x.x.x

Syllabification

Automatically break words into syllables matching specified patterns.

 
Syllables:
  {b, n} a

banana => ba.na.na
banina => #Error!

Resyllabification

Words are automatically broken into syllables again at the end of each named rule.

 
Syllables:
  {b, n} a
rule:
  a => aba / n _ n

banana => ba.na.ba.na

New syllabification

Change syllabification patterns between rules.

 
Syllables:
  {b, n} a
rule:
  a => i / _ n
Syllables:
  {b, n} {a, i}

banana => bi.ni.na

Clearing syllables

Throw away all syllable information for subsequent rules.

 
Syllables:
  explicit
rule:
  a => i / _ n
Syllables:
  clear

ba.na.na => binina

Syllable-level features

These features attach to whole syllables rather than sounds.

 
Feature (syllable) +stress
Diacritic ' (before) [+stress]
Syllables:
  explicit
rule:
  a&[+stress] => e

'ba.na.na => 'be.na.na
ba.'na.na => ba.'ne.na

Syllabification features

Apply syllable-level features to syllables that match certain patterns

 
Feature (syllable) +stress
Diacritic ' (before) [+stress]
Syllables:
  b a => [+stress]
  n a

banana => 'ba.na.na
nabana => na.'ba.na