The Language Creator
– from linguistic questionnaire to grammar, lexicon and texts.

Questionnaire

This questionnaire is the main entry point to the Language Creator, a tool for generating languages by specifying their features.

Throughout this questionnaire, example languages are provided as reference points. The settings shown indicate values that tend to produce results broadly comparable to the named languages, and are intended as guidance rather than precise descriptions of the languages themselves.

Some questions present multiple discrete choices; others use a slider, typically ranging from 0 to 1. For slider-based questions, lower values minimise a feature and higher values maximise it, within a range considered plausible for natural languages.

For convenience, the questionnaire has been pre-filled with random values, so you can simply press Submit, or adjust any of them first. Alternatively, you can choose a starting point:

Once you have completed the questionnaire, click Submit in the corner to generate your language.

Table of contents
  1. Phonology
    1. Consonants
    2. Vowels
    3. Phonotactics, stress and tone
  2. Morphology
    1. Morphological Structures and Strategies
    2. Grammatical categories
  3. Head and Dependent Marking
    1. The sentence
    2. Possessive construction
    3. Adpositions (prepositions, postpositions etc.)
  4. Syntax
    1. Pro-drop
    2. Greenbergian Word order

Phonology

Consonants

Consonants are speech sounds produced with a narrowing or complete closure somewhere in the vocal tract, obstructing airflow to varying degrees. Some languages, like Rotokas, have as few as six consonant phonemes, while others, such as !Xóõ, have over a hundred. Most languages fall somewhere in between, with around 20 to 25 consonants. In this section, we’ll explore first where these sounds are made in the mouth and throat (place of articulation), and then how they are made (manner of articulation).

Places of articulation

The place of articulation describes where in the vocal tract the constriction occurs during consonant production – from the lips at the front to the glottis at the vocal folds.

Each of the following parameters represents a class of consonants defined by its articulatory location.

Question 1: Labials

Labial sounds involve the lips – either both lips (bilabials like /p/ and /m/) or the lower lip against the upper teeth (labiodentals like /f/ and /v/).

Bilabials are nearly universal in human languages, whereas labiodentals are slightly less common but still widespread.

01

Example values:

  • 0.2: Maori, Hawaiian
  • 0.3: TokiPona, Greenlandic
  • 0.4: Spanish, Russian, Japanese
  • 0.5: French, English

Further reading:

Question 2: Dentals

Dental consonants are produced with the tongue against the teeth, while alveolar ones are articulated just behind the teeth at the alveolar ridge.

Many languages include both types, but only a few distinguish them phonemically across the same manner of articulation (e.g. /t̪/ vs /t/).

01

Example values:

  • 0: Maori, Hawaiian, Japanese, French, TokiPona, Greenlandic
  • 0.4: Spanish, Russian
  • 1.0: English

Further reading:

Question 3: Retroflex

Retroflex sounds are articulated with the tongue tip or blade curled back toward the postalveolar region or hard palate. These are common in Indo‑Aryan, Dravidian, Australian Aboriginal, and some East/Southeast Asian languages.

Approximately 20 % of documented languages have at least one retroflex phoneme (e.g. ʈ, ɳ, ɭ), and they tend to occur only in phonological systems with other coronal contrasts.

01

Example values:

  • 0: Maori, Hawaiian, Japanese, Spanish, Russian, French, TokiPona, Greenlandic, English

Further reading:

Question 4: Palatals

Palatal consonants are produced by raising the body of the tongue toward the hard palate. These may include the palatal approximant /j/ as well as palatal stops, nasals, fricatives, or affricates.

The palatal approximant /j/ is extremely common cross-linguistically, while languages with a full series of palatal consonants (e.g. /ɲ, c, ɟ, ç/) are less widespread. Some languages lack primary palatal consonants altogether; others include one or more as part of their core inventory.

01

Example values:

  • 0: Maori, Hawaiian
  • 0.1: TokiPona, Greenlandic, English
  • 0.2: Japanese
  • 0.4: Spanish, French
  • 1.0: Russian

Further reading:

Question 5: Laryngeals

These are consonants articulated using the larynx (voice‑box), including glottal (e.g. /ʔ/, /h/), uvular (e.g. /q/, /χ/), pharyngeal (e.g. /ʕ/, /ħ/) and epiglottal (e.g. /ʡ/) sounds.

Glottal consonants are common cross‑linguistically, uvulars occur with moderate frequency, pharyngeals are typologically rare, and epiglottals are extremely rare.

These places of articulation form an implicational hierarchy: the presence of epiglottals almost always implies pharyngeals, which in turn imply uvulars, and uvulars typically imply glottals.

01

Example values:

  • 0: Russian, Spanish, French, TokiPona
  • 0.1: English
  • 0.2: Maori, Japanese
  • 0.4: Hawaiian
  • 0.5: Greenlandic

Further reading:

Manner of articulation

Manner of articulation refers to how the airstream is shaped or obstructed as it moves through the vocal tract. Different manners produce different types of consonants, such as stops, fricatives or nasals.

Question 6: Nasal stops

Nasal stops are consonants such as /m/, /n/, and /ŋ/, produced by allowing air to escape through the nose while the oral cavity is blocked. Nasal stops are extremely common cross-linguistically and are often among the most stable segments in a phonemic inventory. Some languages, however, lack nasal stops entirely.

01

Example values:

  • 0.1: TokiPona
  • 0.2: Japanese, Hawaiian, Russian, French
  • 0.3: Spanish
  • 0.4: Greenlandic, English
  • 0.6: Maori

Further reading:

Question 7: Fricatives

Fricatives are consonants produced by forcing air through a narrow constriction in the vocal tract, creating turbulent airflow. Common examples include /f/, /θ/, /s/ and /x/. Fricatives are widespread cross-linguistically, though some languages restrict them to a small subset of places or lack them entirely.

01

Example values:

  • 0: Japanese, Hawaiian, TokiPona
  • 0.4: Maori, Russian
  • 0.5: Greenlandic
  • 0.6: English
  • 0.7: Spanish, French

Further reading:

Question 8: Approximants

Approximants are consonants produced with a narrowing of the vocal tract wide enough to avoid turbulence. They are vowel-like sounds such as /w/, /j/, or /ʋ/, distinct from fricatives.

01

Example values:

  • 0.2: Maori, Hawaiian, Russian, TokiPona, Greenlandic, English
  • 0.4: Japanese, Spanish, French

Further reading:

Question 9: Sibilants

Sibilants are a class of fricative consonants characterised by a high-pitched hissing or hushing sound, typically produced by directing a narrow stream of air against the teeth. Common sibilants include /s/, /z/, /ʃ/ and /ʒ/. Some languages also have other types, such as retroflex sibilants.

Sibilants are frequent cross-linguistically, but their number and role in the phonemic system vary considerably between languages.

01

Example values:

  • 0.0: Maori, Hawaiian
  • 0.1: TokiPona, Greenlandic
  • 0.2: Spanish, Japanese
  • 0.3: French, English
  • 1.0: Russian

Further reading:

Question 10: Voicing

Voicing refers to whether the vocal folds vibrate during the production of a consonant. In some languages, voicing is phonemically contrastive, as in /p/ vs /b/ or /t/ vs /d/, allowing words to be distinguished by the presence or absence of vocal fold vibration.

Other languages lack phonemic voicing contrasts altogether, while many fall between these extremes, restricting voicing contrasts to certain consonants or relying primarily on other cues.

01

Example values:

  • 0: Maori, Hawaiian, TokiPona, Greenlandic
  • 0.4: Spanish
  • 0.6: French
  • 0.7: English
  • 1: Japanese, Russian

Further reading:

Question 11: Aspiration

Aspiration refers to a period of audible breath following the release of a consonant, most commonly a stop. In some languages, aspiration is phonemically contrastive, distinguishing otherwise similar sounds such as /p/ vs /pʰ/.

Many languages lack phonemic aspiration altogether, while others restrict it to particular consonants or environments, or use it alongside other cues such as voicing or length. In some languages, including Mandarin Chinese, contrasts often written ⟨b⟩ vs ⟨p⟩ reflect differences in aspiration rather than voicing.

01

Example values:

  • 0: Maori, Spanish, Hawaiian, Japanese, Russian, French, TokiPona, Greenlandic, English

Further reading:

Question 12: Affricates

Affricates are consonants that begin as stops and release into fricatives, functioning as single phonemic units rather than as consonant clusters. Typical examples include /ts/, /tʃ/, /dz/, and /dʒ/. Some languages lack affricates entirely, while others use a small number alongside plain stops and fricatives. In languages with richer affricate inventories, these sounds may form parallel series across several places of articulation.

01

Example values:

  • 0: Maori, Hawaiian, Japanese, French, TokiPona, Greenlandic
  • 0.2: Spanish
  • 0.3: English
  • 0.4: Russian

Further reading:

Question 13: Liquids

How many liquid consonants should the language have? Liquids include laterals like /l/ and various kinds of /r/ sounds (rhotics), which are typically more sonorous and fluid than other consonants.

01

Example values:

  • 0.1: French, TokiPona, Greenlandic
  • 0.2: Maori, Hawaiian, Japanese
  • 0.5: English
  • 0.6: Russian
  • 0.8: Spanish

Further reading:

Question 14: Ejectives

Should ejective consonants be part of the language? These are voiceless consonants produced with a burst of air from the glottis, found in languages like Amharic or Quechua.

01

Example values:

  • 0: Spanish, Russian, Maori, Hawaiian, Japanese, French, TokiPona, Greenlandic, English

Further reading:

Question 15: Labialisation

Should labialised consonants be part of the language?

01

Example values:

  • 0: English, French, Hawaiian, Hawaiian, Japanese, Maori, Maori, Russian, Spanish, TokiPona, Greenlandic

Further reading:

Question 16: Pharyngealisation

Should pharyngealised consonants be part of the language?

01

Example values:

  • 0: English, French, Hawaiian, Hawaiian, Japanese, Maori, Maori, Russian, Spanish, TokiPona, Greenlandic

Further reading:

Question 17: Implosives

Should the language include implosive consonants? These are voiced stops made by drawing air inward while voicing, found in languages like Sindhi or Swahili.

01

Example values:

  • 0: Hawaiian, Hawaiian, Japanese, Maori, Maori, Russian, Spanish, French, TokiPona, Greenlandic, English

Further reading:

Question 18: Clicks

Should the language have click consonants? These are rare and exotic from a global perspective, but richly used in southern African languages like Xhosa and !Xóõ.

01

Example values:

  • 0: Spanish, Maori, Hawaiian, Japanese, Russian, French, TokiPona, Greenlandic, English

Further reading:

Vowels

Vowels are produced with an open vocal tract and are classified by height, backness, rounding and nasality. This section shapes the vocalic system of the language.

Oral Vowel Heights

The vertical tongue position – high, mid, low – defines vowel height. Languages typically have between two and five distinct heights.

Question 19: Vowels heights

How many vowel heights should the language distinguish?

24

Example values:

  • 2: Greenlandic
  • 3: Maori, Russian, Hawaiian, Japanese, Spanish, TokiPona
  • 4: French, English

Further reading:

Question 20: Nasal vowels

Should nasal vowels occur in the language, and if so, how many height distinctions should they have? 0 means no nasal vowels; 1 means nasal vowels mirror the full height range of oral vowels.

01

Example values:

  • 0: Greenlandic, Maori, Russian, Hawaiian, Japanese, Spanish, TokiPona, English
  • .4: French

Further reading:

Question 21: Vowel length

Should the language distinguish long and short vowels? 0 means no vowel length, 1 means vowel length is distinctive everywhere, and values inbetween mean length will be distinctive in some positions.

01

Example values:

  • 0.0: Russian, Spanish, French, TokiPona, English
  • 1: Greenlandic, Maori, Hawaiian, Japanese

Further reading:

Question 22: Backness

How many degrees of vowel backness (front, central, back) should the language employ? A value of 1 represents a vertical vowel system, where vowel quality is shaped by surrounding consonants rather than being phonemically distinct. Such systems are rare but attested – for instance in Abkhaz, Marshallese, and Kabardian (and according to some analyses, Mandarin Chinese) – where a small number of centralised vowel phonemes surface as front or back depending on context. A value of 2 means the system distinguishes only front and back vowels, while 3 allows for front, central, and back as distinct categories.

13

Example values:

  • 2: Greenlandic, Maori, Russian, Hawaiian, Japanese, Spanish, French, TokiPona, English

Further reading:

Question 23: Squareness

How geometrically square should the vowel system be? A square system balances high and low vowels across frontness positions, forming a roughly rectangular shape in vowel space. A triangular system crowds vowels in the high region and typically has only one or two low vowels. A value of 1 yields a triangular system, 3 gives a square system, and 2 is intermediate.

13

Example values:

  • 2: Maori, Russian, Greenlandic, Hawaiian, Japanese, Spanish, French, TokiPona
  • 3: English

Further reading:

Question 24: Reduced vowels

How many vowel heights should occur in reduced syllables (0 = schwa only, 1 = full set)?

01

Example values:

  • .4: English, Russian
  • 1: French
  • 1: Maori, Hawaiian, Japanese, Spanish, TokiPona, Greenlandic

Further reading:

Question 25: Empty nuclei (ghost schwa)

Many languages allow weak or “ghost” nuclei that may not surface as audible vowels. These often appear as /ᵊ/ in underlying representations but delete in the output.

At the lowest setting, every syllable must contain a full vowel. As the value increases, weak nuclei become more frequent, and eventually words may surface without any vowel at all.

01

Example values:

  • 0.0: Greenlandic, English, French, Hawaiian, Hawaiian, Maori, Maori, Russian, Spanish, French, English
  • 0.2: Japanese

Further reading:

Question 26: Roundedness

Should vowel roundedness be phonemically distinctive?

the language distinguishes between front rounded and unrounded vowels (/i/ vs. /y/)– e.g., French
the language distinguishes between back rounded and unrounded vowels (/u/ vs. /ɯ/)
the language should not use roundedness– e.g., Maori, Hawaiian, Greenlandic, Japanese, Spanish, Russian, TokiPona, English

Further reading:

Question 27: Vowel harmony

Should the language use vowel harmony?

No– e.g., Maori, Russian, Hawaiian, Japanese, Spanish, Greenlandic, French, English
Yes

Further reading:

Phonotactics, stress and tone

Beyond individual consonants and vowels, languages differ in how they structure syllables, place emphasis, and use pitch or vowel quality to convey grammatical or lexical meaning. This section covers these broader phonological patterns, which often shape a language’s overall sound and rhythm.

Syllable complexity

Many languages vary markedly in how elaborate their syllable structures can be. A universally common pattern is the CV (consonant–vowel) syllable, typical of languages like Polynesian or Hawaiian. In contrast, other languages permit clusters of consonants in onsets and codas to varying degrees. This range – from strictly simple syllables to highly complex ones – can influence phonological shape, rhythm, and learnability. Use the scale below to determine where the language fits in this typological spectrum.

Question 28: Diphthongs

Should there be diphthongs?

01

Example values:

  • 0: Greenlandic, Maori, Russian, Hawaiian, Japanese, Spanish, TokiPona, French
  • .4: English
Question 29: Coda complexity (relative to onsets)

Codas vary widely across languages. Some have no codas at all, others allow simple codas only, and still others permit codas as complex or even more complex than onsets.

This slider sets how heavy codas can be compared to the onset cluster inventory.

01

Example values:

  • 0.0: Hawaiian, Japanese
  • 0.6: English
  • 0.8: Russian

Further reading:

Stress and tone

Stress refers to the relative prominence of syllables within a word. It may be absent, predictable (e.g. always on the first syllable), or contrastive (e.g. shifting meaning when placed differently). Tone involves using pitch to distinguish lexical or grammatical meaning. Some languages use only two tones; others use four or six. Non-tonal languages still use pitch, but only for intonation or emphasis. If both stress and tone are selected, the language will restrict tonal contrasts to the stressed syllable, creating a pitch-accent system similar to Swedish or Serbian. If phonation contrast is also selected, this may resemble systems such as Danish, where pitch and glottalisation interact.

Question 30: Stress

Should the language use stress?

01

Example values:

  • 0: French, Greenlandic, Japanese, Maori, Hawaiian
  • 1: English
Question 31: Tones

How many lexical tones should the language have? 0 means it is non-tonal; values from 2 to 6 reflect the number of distinct pitch patterns used to distinguish meaning.

06

Example values:

  • 0: French, Greenlandic, English, Maori, Hawaiian, Japanese
Question 32: Phonation contrast on tones

If the language has lexical tone, should some tones also be distinguished by phonation (e.g. glottalisation or pharyngealisation), creating contrasts beyond pitch alone? A well-known example of such a feature is the Danish stød. If the language doesn't have tones, this will attach to the vowels in the same way as vowel length.

No– e.g., English, Russian
Yes

Morphology

Morphological Structures and Strategies

Morphology studies how words are built from morphemes. This section asks which morphological tools – affixes, clitics, fusion and so on – the language will exploit.

Affix position and usage

This section controls how bound grammatical elements behave in the language – including where they appear, how tightly they bind, and whether they are required for well-formedness. The parameters here influence whether the language uses prefixes or suffixes, whether it prefers affixes or clitics, and whether grammatical categories like number or tense must always be overtly marked.

Question 33: Before or after

This setting controls the position of grammatical markers relative to their host. At low values, most affixes and clitics follow their head (suffixing, enclitic languages). At high values, they precede the head (prefixing, proclitic languages). This applies across categories: e.g., noun case markers, verb agreement markers, tense/aspect, etc.

01

Example values:

  • 0: Japanese, Greenlandic
  • 0.2: English
  • 0.4: French
  • 0.8: Maori, Hawaiian, TokiPona
Question 34: Clitics

This setting controls how commonly clitics are used in the language. Clitics are morphologically simpler than affixes – they are phonologically bound to a host, but behave more like independent words syntactically. High values produce languages with many clitics (e.g., for tense, mood, negation, discourse particles, or pronouns). Low values favour true affixes (i.e., tightly bound and ordered morphemes).

01

Example values:

  • 0: Greenlandic
  • 0.6: French, English
  • 0.8: Maori, Hawaiian
  • 1.0: TokiPona, Japanese
Question 35: Is marking optional or obligatory?

This setting determines how obligatory morphological marking is in the language. At low values, key grammatical features like number, case, tense, or mood must always be overtly marked. At high values, affixes are used only when needed to disambiguate, and the default is left unmarked. For example, in Latin, servus (“slave”) must always end in -us – there’s no bare serv. This reflects high obligatoriness. In contrast, in English, cat is unmarked for number, and -s is only added for plural – marking is optional and asymmetric. This parameter is especially important when generating defaults: if set to 0.0, every grammatical value must have a morpheme. If set to 1.0, the most frequent or “unmarked” value will appear with no overt marking.

01

Example values:

  • 1.0: Maori, Hawaiian
  • 0.6: French, English, Japanese, Greenlandic

Fusion and complexity

Highly agglutinative systems (e.g. Turkish) string many clear morphemes, while fusional systems (e.g. Latin) blend information into single endings.

Question 36: Fusion

How much morphological fusion should affixes exhibit?

01

Example values:

  • 0: Maori, Hawaiian
  • 0.2: Japanese
  • 0.4: French, English

Polysynthesis and Marking Strategies

Polysynthesis refers to the tendency of a language to pack large amounts of meaning – including entire clauses – into single words. But there are different ways to do this. Some languages, like Greenlandic and Turkish, use long sequences of suffixes to add detail to a basic root (e.g. “house‑have‑3sg” for “he has a house”), while others, like Mohawk or Chukchi, treat the verb as the central grammatical hub and incorporate all its arguments directly into it – including the object noun. This section lets you set two related parameters. The first controls how much the language uses affixes to build complex words (affix density). The second controls whether and how strongly the language incorporates subjects, objects, and other arguments into the verb (argument incorporation). Both strategies can result in long words, but the underlying grammar is quite different.

Question 37: Affix density

This setting controls how complex individual words are. A high affix density means that words often contain many suffixes or prefixes that express things like tense, mood, direction, negation, causation, evidentiality, politeness, etc. These are often built up in a templatic way, with strict ordering of elements. Affixes here may be inflectional (e.g. -ed, -s) or derivational (e.g. -er, re-).

High affix density does not imply that arguments (like subject and object) are inside the verb – only that words tend to be long and internally structured.

01

Example values:

  • 0.0: TokiPona, English
  • 0.2: Maori, Hawaiian, Japanese, French
  • 1.0: Greenlandic
Question 38: Incorporation

This setting determines how much the language builds verbs that encode their arguments directly. In languages with high incorporation, a verb like “see” may include markers for “I” and “him” and absorb the object itself – “I-him-fish-see” for “I saw the fish”. (Absorbing the object is called noun incorporation.) This makes the verb into a mini-sentence, and often means that independent noun phrases are optional or ungrammatical. In Baker’s sense, a fully polysynthetic language is one where every argument of a predicate (subject, object, etc.) must be either agreed with or incorporated into the verb. This is what distinguishes languages like Mohawk from languages like Greenlandic or Turkish: even though all three form long words, only some treat the verb as the grammatical centre of gravity.

At high values, the language will tend to “swallow” its arguments, leaving the verb to carry the load. Object nouns may be turned into verb prefixes or incorporated roots, and subject pronouns may be replaced with agreement markers. If affix density is also high, words will become very dense and clause-like. If affix density is low but incorporation is high, you’ll get polypersonal agreement without derivational layering – as in some Iroquoian languages.

01

Example values:

  • 0: Japanese, Maori, Hawaiian, French, Greenlandic, English

Grammatical categories

Grammatical categories – number, gender, tense and more – encode abstract contrasts that shape clause structure.

Number

Number distinguishes quantities such as singular, dual or plural.

Question 39: Number

Should the language have grammatical number?

it should not have number as a grammatical category– e.g., Maori, Hawaiian, Japanese, TokiPona
the language distinguishes between singular and plural– e.g., English, French, English
the language employs singular, dual and plural

Gender

Grammatical gender or noun‑class systems partition nouns into sets that trigger agreement.

Question 40: Gender

Should the language have grammatical gender?

it should not have gender as a grammatical category– e.g., Maori, Hawaiian, Japanese, TokiPona, English
masc/fem– e.g., French, Spanish
anim/inanim
the language employs masculine, feminine and neuter– e.g., Russian
the language has four genders
the language has noun classes

Tense, Aspect and Mood (TAM)

Tense locates events in time, aspect profiles their internal structure, and mood expresses speaker attitude or modality.

Articles and definiteness marking

Languages often distinguish between definite and indefinite noun phrases – roughly corresponding to whether the referent is assumed to be identifiable by the listener. This distinction may be marked overtly by dedicated words (such as articles), through affixes, by word order, or may not be marked at all. Where articles exist, they are typically used to encode definiteness (“the”) or indefiniteness (“a”), though their distribution and semantics vary cross-linguistically.

Question 41: Definite articles

Does the language use a dedicated marker for definiteness, such as a definite article or affix?

Yes– e.g., Maori, French, Spanish, Hawaiian, English, English
No– e.g., Russian, TokiPona, Japanese

Further reading:

Question 42: Indefinite articles

Does the language use a dedicated marker for indefiniteness, such as an indefinite article or affix?

Yes– e.g., Spanish, English, English
No– e.g., Maori, Hawaiian, French, TokiPona, Japanese, Russian

Further reading:

Should there be adjectives? (Or just nouns and verbs?)

Languages may treat property words as a separate class (adjectives) or fold them into verbs or nouns.

Question 43: Adjectives

Languages may treat property words as a separate class (adjectives) or fold them into verbs or nouns.

Adjectives are verbs– e.g., Maori, Hawaiian, Japanese
Adjectives are nouns
Adjectives are separate from both– e.g., English, French, English

Head and Dependent Marking

The sentence

Languages differ in whether grammatical relations are marked on the head (verb) or on the dependents (noun phrases). This section controls how subject and object relations are marked on noun phrases, including which alignment system is used and how broadly it applies.

NP alignment: How subject and object are marked on noun phrases

Different languages mark subject and object noun phrases in different ways. Some use a nominative–accusative system, marking direct objects differently from subjects. Others use an ergative–absolutive system, marking transitive subjects differently. Some languages use active–stative alignment, where intransitive subjects are marked like agents or patients depending on semantics. This setting chooses the type of alignment used on noun phrases, and how widely this marking is applied across the hierarchy of noun types.

Question 44: NP alignment type

What kind of alignment should noun phrases use?

noun phrases are not marked for grammatical roles (neutral alignment)– e.g., TokiPona
nominative–accusative: subjects are treated alike, objects are marked– e.g., Japanese, French, English
ergative–absolutive: objects and intransitive subjects are treated alike, agents are marked– e.g., Maori, Hawaiian
active–stative: intransitive subjects are marked differently depending on agency or volition
split alignment: ergative alignment applies in the past/perfective tense only

Verb alignment: How subject and object are marked on the verb

Many languages use verbal affixes or agreement markers to show who is doing what to whom. This may follow different alignment strategies: for example, subjects of intransitive and transitive verbs may be marked the same (accusative alignment), or intransitive subjects may be treated like transitive objects (ergative alignment). Some languages split intransitive subjects based on semantic roles like volition or control (active–stative), or use person hierarchies instead of grammatical roles. This setting determines which **alignment pattern** the verb follows when it indexes or agrees with arguments. The degree to which the verb actually marks its arguments – from optional agreement to full incorporation – is controlled by the separate incorporation setting elsewhere in the questionnaire.

Possessive construction

Possession can be marked on the possessor, the possessed noun, both or neither.

Locus of Marking in Possessive Noun Phrases

Should the possessor carry marking (e.g. a genitive case or adposition)?

Question 45: Marking the owner

Mark the possessor?

it should mark the possessor– e.g., Japanese, French, TokiPona, English
it should not mark the possessor

Marking of owned (possessive)

Should the possessed noun itself bear possessive morphology?

Question 46: Marking the possessed

Mark the possessed item?

the language marks the possessed noun ("the man his house")
it should not mark the possessed noun– e.g., TokiPona, Japanese, French, English

Adpositions (prepositions, postpositions etc.)

Adpositions are words or morphemes that express relationships between a noun and another element in the clause — typically direction, location, possession or instrument. They can appear as prepositions (before the noun), postpositions (after), or in more complex forms. Some languages mark these relationships instead with case, clitics, or verbs.

Marking

Languages vary in how they mark the relationship between the noun and the adposition. Some mark the head of the phrase (the adposition), others the dependent (the noun), and some do both.

  • Head‑marking: The adposition shows agreement with the noun or pronoun it governs. For example, Scottish Gaelic uses different forms like agam (“at me”) and agad (“at you”).
  • Dependent‑marking: The noun shows case or other morphology governed by the adposition. For instance, in German, the preposition mit (“with”) requires the dative case: mit dem Mann (“with the man”).

Question 47: Head-marking on PPs

Should adpositions in the language agree with their noun complement? This is known as head-marking. It often results in different adpositional forms for different persons or numbers of the noun.

yes
no– e.g., TokiPona, French, Japanese, English
Question 48: Dependent-marking on PPs

Should adpositions govern case marking on their noun complements? This is known as dependent-marking. It is common in languages with rich case systems, where the noun form reflects the role imposed by the adposition.

yes
no– e.g., TokiPona, English

Prepositions or postpositions

Adpositions may come before or after their noun complement. English uses prepositions like “on the table”, whereas Japanese uses postpositions like ie ni (“to the house”). This setting determines the word order of adpositional phrases.

Question 49: Position of adpositions

Should the language use prepositions (before the noun) or postpositions (after the noun)?

It should use prepositions– e.g., English, TokiPona, French, Spanish
It should use postpositions– e.g., Japanese

Complex adpositional phrases

Some languages build complex spatial expressions from simpler parts. For example, instead of having a single word for “inside”, they might say “in the inside of the house”. This option models whether your language prefers to deconstruct such concepts into multiple elements, allowing for more regular but less lexicalised structures.

Question 50: Decomposing complex adpositions

Should the language prefer to build complex adpositional meanings (like “into” or “from inside”) using combinations of simpler morphemes such as “in” + “interior”, rather than treating them as distinct adpositions?

yes– e.g., Japanese, TokiPona
no– e.g., English, English

Syntax

Pro-drop

Pro-drop

Question 51: Prodrop

This setting controls whether subject and object pronouns may be omitted when recoverable from context – a phenomenon known as pro-drop or zero anaphora.

01

Example values:

  • 0: TokiPona, English
  • 1: Japanese

Greenbergian Word order

Constituent order (S, V, O) correlates with many other syntactic properties.

Order of verb, subject and object

Pick the dominant S‑V‑O ordering pattern for a simple transitive clause.

Question 52: S, V and O

Which basic order should the language adopt (S = subject, V = verb, O = object)?

it should use OSV
it should use OVS
it should use SOV– e.g., Japanese
it should use SVO– e.g., French, TokiPona, English
it should use VOS
it should use VSO– e.g., Maori, Hawaiian

NDem or DemN

Should demonstratives come before or after the noun?

Question 53: Demonstrative order

Choose demonstrative order.

it should place demonstratives before the noun– e.g., French, Japanese, English
it should place demonstratives after the noun

AN or NA

Should adjectives precede or follow the noun?

Question 54: Adjective order

Choose adjective order.

Adjective + Noun– e.g., Japanese, English
Noun + Adjective– e.g., French

NGen or GenN

Should the possessor precede or follow the possessed noun?

Question 55: Genitive order

Choose genitive order.

it should place the possessor after the possessed– e.g., TokiPona, French, Hawaiian, Maori
it should place the possessor before the possessed– e.g., Japanese, English

NNum or NumN

Should numerals come before or after the noun?

Question 56: Numeral order

Choose numeral order.

Numeral + Noun– e.g., French, English
Noun + Numeral– e.g., Japanese

Relative and complement clauses

Many languages allow whole clauses to be embedded inside a larger sentence. This happens when a clause modifies a noun (a relative clause) or when it functions as the object of a verb such as “know”, “say” or “think” (a complement clause). Different languages use different strategies for forming these structures.

Question 57: Relative clause strategy

Languages differ in how they form relative clauses. In many systems the clause follows the head noun and may contain a dedicated relativiser (“I know the man who sings”). Other languages place the clause before the noun without any relativiser, so the entire clause behaves like an adjectival modifier (“I know [sings man]”). A third type keeps the head noun inside the modifying clause, so that the whole clause functions as a noun phrase; this type uses the same mechanisms that the language employs for embedding or nominalising clauses (see the following question). Finally, some languages use a correlative strategy, pairing a relative clause with a demonstrative element in the main clause.

Postnominal relative clause (after the noun, often with a relative pronoun) – e.g., French, English
Prenominal relative clause (before the noun, no relative pronoun) – e.g., Japanese
Embedded/head-internal strategy.
Correlative strategy (a relative clause introduced by a relativiser, paired with a demonstrative in the main clause)

Further reading:

Question 58: Complement clause strategy

Complement clauses occur when an entire clause functions as the object of a verb, as in “I know that he comes”. Languages approach this in several ways: by using a dedicated complementiser (like “that” in English), by turning the clause into a nominal expression (“I know his coming”), or by marking the finite verb with subordinating morphology (as in Basque). Which strategy should this language use?

Complementiser (dedicated particle introducing the clause) – e.g., English
Nominalisation (the clause behaves as a noun phrase) – e.g., Japanese
Subordinating verbal morphology (finite clause with subordinating suffix)

Further reading:

Writing system

This setting determines the script used to display example texts and lexical forms. The choice affects presentation only and does not constrain the underlying phonology.

Question 59: Script

This setting determines the script used for displaying example texts and lexical forms. The choice does not constrain the underlying phonology, and some scripts may require adaptations or approximations if the language’s sound system exceeds their traditional coverage.

If no script is selected, the language will be presented using IPA only.

Scripts marked [beta] are partially implemented. They may currently lack support for certain phonological features (such as clicks or front rounded vowels), which can lead to IPA symbols popping up unexpectedly.

Latin– e.g., Maori, Hawaiian, Japanese, Spanish, French, TokiPona, Greenlandic, English
Cyrillic [beta]– e.g., Russian
Devanagari [beta]
Arabic [beta]
None (only IPA will be supplied)
Last Question: Seed (optional)

If you enter a value here, the same underlying language will be generated each time. This can be useful if you wish to experiment with changes (for example in the orthography) while keeping the structural core of the language constant. If left blank, a new random language is generated each time.