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

Language Creation Questionnaire

The simplest way to use the Language Creator is to click Create. The questionnaire is already filled in with random choices, so you can generate a complete language immediately.

Once the grammar has appeared, you can use Back to return to this questionnaire. Clicking Create again will generate another language from the same settings. Since generation normally takes less than a second, it is often worth doing this a few times until you get a result you like. There is no account, no fee, no quota and no cookies.

You can also adjust one or two settings before generating again. If you want a completely fresh set of random choices, reload this page. You can also use one of the reference languages below as your starting point:

If you find completing the questionnaire daunting, you can use the AI mode instead.

A few notes on the questionnaire itself: Throughout it, 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.

Table of contents
  1. Phonology
    1. Consonant inventories
    2. Vowel systems
    3. Syllable structure and prosody
  2. Morphology
    1. Morphological structure
    2. Grammatical categories
  3. Alignment and marking
    1. Clause alignment
    2. Possession
    3. Adpositions and adpositional phrases
  4. Syntax and word order
    1. Argument expression
    2. Basic word order
  5. Lexicon and writing
    1. Lexicon and writing

Phonology

Consonant inventories

Consonants are speech sounds made by narrowing or blocking the vocal tract. Languages differ both in where consonants are made and in how the airflow is shaped. This section sets the main consonant inventory: places of articulation, manners of articulation and special consonant types. These choices strongly affect the sound of the generated language, because they determine which consonants can appear in roots, affixes and example texts.

Places of articulation

Place of articulation describes where a consonant is made: at the lips, teeth, alveolar ridge, palate, uvula, pharynx, glottis and so on. The following sliders set how strongly different regions of the vocal tract are represented in the consonant inventory.

Question 1: Labials

Labial consonants involve the lips: both lips for sounds such as /p/ and /m/, or lower lip against upper teeth for sounds such as /f/ and /v/. Low values keep this area sparse; high values give the language a broader set of labial and labiodental contrasts.

01

Example values:

  • 0: Oneida
  • 0.1: Finnish, Guarani, Hawaiian, Japanese, Tlingit, Wajarri, TokiPona, Greenlandic, Pirahã, Navajo
  • 0.2: Korean
  • 0.4: Archi
  • 0.5: Arabic, Danish, Esperanto, Hindi, Mandarin, Maori, Russian, Spanish, VietnameseS, Thai, Bengali, Hausa, Nama, Basque, !Xóõ
  • 0.6: Irish, Ubykh, VietnameseN, English, Portuguese, Italian, Turkish, Persian, Telugu, Indonesian, Tagalog, Swahili, Malagasy
  • 0.7: German
  • 0.8: Ewe, Georgian
  • 1.0: French

Further reading:

Question 2: Dentals

Dental consonants are made with the tongue against or near the teeth, while alveolars are made just behind them. This setting controls whether the language treats dental articulation as a distinct part of the consonant system, rather than folding it into a general coronal series.

01

Example values:

  • 0: Archi, Danish, Esperanto, Finnish, French, Guarani, Hawaiian, Hindi, Irish, Japanese, Mandarin, Maori, Russian, Tlingit, Ubykh, VietnameseN, VietnameseS, TokiPona, Greenlandic, Portuguese, German, Korean, Italian, Turkish, Thai, Persian, Bengali, Telugu, Indonesian, Tagalog, Hausa, Georgian, Basque, Malagasy, Pirahã, Navajo, Oneida
  • 0.1: Nama
  • 0.2: Spanish
  • 0.3: Arabic, English, Swahili
  • 0.8: Ewe, !Xóõ
  • 1.0: Wajarri

Further reading:

Question 3: Retroflexes

Retroflex sounds are made with the tongue tip or blade curled back towards the postalveolar or palatal region. Low values omit them; higher values allow increasingly systematic retroflex contrasts such as stops, nasals, laterals or sibilants.

01

Example values:

  • 0: Arabic, Archi, Danish, Esperanto, Ewe, Finnish, French, Guarani, Hawaiian, Irish, Japanese, Mandarin, Maori, Russian, Spanish, Tlingit, VietnameseN, TokiPona, Greenlandic, English, Portuguese, German, Korean, Italian, Turkish, Thai, Persian, Indonesian, Tagalog, Hausa, Nama, Georgian, Basque, Malagasy, Pirahã, Navajo, !Xóõ, Oneida, Swahili
  • 0.5: Ubykh, VietnameseS
  • 0.6: Wajarri
  • 0.9: Telugu
  • 1.0: Hindi, Bengali

Further reading:

Question 4: Palatals

Palatal consonants are made by raising the body of the tongue towards the hard palate. This setting controls whether the language has only little or no palatal material, a few sounds such as /j/ or /ɲ/, or a wider palatal and palatalised subsystem.

01

Example values:

  • 0: Hawaiian, Mandarin, Maori, Georgian, Pirahã
  • 0.1: Arabic, Archi, Esperanto, Ewe, Finnish, Hindi, Japanese, Tlingit, TokiPona, Greenlandic, English, German, Persian, Bengali, Hausa, Malagasy, Navajo, Oneida
  • 0.3: Danish
  • 0.4: French, Guarani, VietnameseN, Telugu, Indonesian, Tagalog, Swahili
  • 0.5: Spanish, Wajarri, Portuguese, Italian
  • 0.6: Korean, Thai
  • 0.7: Ubykh, VietnameseS, Turkish, Nama, Basque, !Xóõ
  • 1.0: Irish, Russian

Further reading:

Question 5: Uvular and pharyngeal consonants

Behind the velum the mouth still has room for several more places of articulation. Uvular consonants are made against the uvula — the little flap at the back of the soft palate — and include /q/, a k pronounced as far back as possible, and /ʁ/, the throaty r of French and German. Pharyngeal consonants are made further down still, by pulling the root of the tongue back into the throat; they are characteristic of Arabic and the languages of the Caucasus. Low values give a language just a uvular stop or two; higher values add fricatives, then pharyngeals, and at the very top the rare epiglottal consonants of languages like Archi.

01

Example values:

  • 0: Danish, Esperanto, Ewe, Finnish, French, Guarani, Hawaiian, Hindi, Irish, Japanese, Mandarin, Maori, Russian, Spanish, VietnameseN, VietnameseS, Wajarri, TokiPona, English, Portuguese, German, Korean, Italian, Turkish, Thai, Bengali, Telugu, Indonesian, Tagalog, Hausa, Swahili, Nama, Basque, Malagasy, Pirahã, Navajo, Oneida
  • 0.2: Georgian, Persian
  • 0.3: Tlingit
  • 0.4: Ubykh, Greenlandic
  • 0.6: Arabic
  • 0.7: !Xóõ
  • 0.9: Archi

Further reading:

Question 6: Glottal consonants

Glottal consonants are made right down in the larynx, at the vocal folds themselves. There are only two in common use: /h/, a breath of air with no obstruction higher in the mouth, and /ʔ/, the glottal stop — the catch in the middle of uh-oh, or the sound some English speakers use in place of the t in bottle. The two are independent: plenty of languages have one without the other. A third, /ɦ/, is a voiced /h/ — the h of Czech hrad or Dutch huis. It is not a separate choice here: where a language asks for /h/, the generator decides whether that comes out as /h/, as /ɦ/, or as both in contrast, depending on how the rest of the phonology treats voicing. Languages with a breathy series, like Hindi or Zulu, reach /ɦ/ most often.

the language has no glottal consonants at all– e.g., Basque, French, Greenlandic, Italian, Portuguese, Russian, Spanish, TokiPona, Wajarri
the language has a glottal fricative /h/ (which the generator may realise as voiced /ɦ/, or both, in the right kind of phonology)– e.g., Bengali, Danish, English, Esperanto, Ewe, Finnish, Georgian, German, Hindi, Irish, Japanese, Korean, Malagasy, Mandarin, Maori, Swahili, Telugu, Ubykh, VietnameseN, VietnameseS, Turkish
the language has a glottal fricative /h/ and the glottal stop /ʔ/– e.g., !Xóõ, Arabic, Archi, Guarani, Hausa, Hawaiian, Indonesian, Nama, Navajo, Oneida, Persian, Pirahã, Thai, Tlingit, Tagalog
the language has the glottal stop /ʔ/ but no /h/

Further reading:

Manner and secondary articulation

Manner of articulation describes how the airflow is shaped: stopped, narrowed, released in stages, nasalised, voiced, aspirated or modified by secondary articulations. These settings determine which broad kinds of consonants the language may use.

Question 7: Nasal stops

Nasal stops such as /m/, /n/ and /ŋ/ are made with oral closure while air escapes through the nose. Most languages have at least some nasals; this setting controls how many places of articulation distinguish nasal stops.

01

Example values:

  • 0.0: Ewe, Pirahã
  • 0.1: Arabic, Archi, Esperanto, Hawaiian, Japanese, Russian, Tlingit, Ubykh, TokiPona, Turkish, Persian, Hausa, Nama, Georgian, Malagasy, Navajo, Oneida
  • 0.2: French, Spanish, Italian, Basque
  • 0.3: Danish, Finnish, Guarani, Hindi, Irish, Mandarin, Maori, VietnameseN, VietnameseS, English, Portuguese, German, Korean, Thai, Bengali, Indonesian, Tagalog, Swahili, !Xóõ
  • 0.4: Greenlandic
  • 0.5: Wajarri, Telugu

Further reading:

Question 8: Fricatives

Fricatives are made by forcing air through a narrow constriction, producing turbulent airflow. Low values reduce or remove them; high values give the language more fricatives across more places of articulation.

01

Example values:

  • 0: Finnish, Guarani, Hawaiian, Japanese, Wajarri, TokiPona, Korean, Bengali, Pirahã, Oneida
  • 0.1: Georgian, !Xóõ
  • 0.2: Danish, Hindi, Maori, Russian, Thai, Indonesian, Hausa
  • 0.3: French, Italian, Turkish, Persian, Telugu, Malagasy
  • 0.4: Tagalog
  • 0.5: Esperanto, Mandarin, Spanish, VietnameseS, English, Portuguese, German, Nama, Basque
  • 0.7: Irish, VietnameseN, Swahili, Navajo
  • 0.9: Archi, Tlingit, Ubykh, Greenlandic
  • 1.0: Arabic, Ewe

Further reading:

Question 9: Approximants

Approximants are vowel-like consonants made with a narrowing that is not tight enough to create frication. This setting controls the presence of sounds such as /w/, /j/ and other smooth sonorant-like consonants.

01

Example values:

  • 0: Pirahã
  • 0.1: Guarani, Tlingit, German, Korean, Thai, Persian, Indonesian, Tagalog, Hausa, Swahili, Nama, Georgian, Malagasy, Navajo, !Xóõ
  • 0.2: Arabic, Archi, Danish, Esperanto, Ewe, Finnish, Hawaiian, Hindi, Irish, Japanese, Maori, Russian, Ubykh, VietnameseN, VietnameseS, TokiPona, Greenlandic, English
  • 0.3: Bengali, Oneida
  • 0.5: Telugu
  • 0.6: Mandarin
  • 0.7: Portuguese, Italian, Turkish, Basque
  • 0.9: French, Spanish, Wajarri

Further reading:

Question 10: Sibilants

Sibilants are high-pitched fricatives such as /s/, /z/, /ʃ/ and /ʒ/. This setting controls how central hissing and hushing sounds are in the consonant inventory, from absent to richly contrastive.

01

Example values:

  • 0.0: Hawaiian, Maori, Wajarri
  • 0.1: Finnish, Hindi, Irish, Japanese, Spanish, VietnameseN, TokiPona, Greenlandic, Bengali, Pirahã, !Xóõ, Oneida
  • 0.2: Archi, Mandarin, Hausa, Nama, Basque
  • 0.4: Ewe, Telugu, Malagasy
  • 0.6: Korean, Thai
  • 0.8: Arabic, Esperanto, French, Guarani, Russian, Tlingit, VietnameseS, English, Portuguese, German, Italian, Turkish, Persian, Indonesian, Tagalog, Swahili, Georgian, Navajo
  • 0.9: Danish
  • 1.0: Ubykh

Further reading:

Question 11: Voicing

Voicing concerns whether the vocal folds vibrate during a consonant. This setting controls whether the language distinguishes pairs such as /p/ and /b/, and how widely such voiced/voiceless contrasts apply.

01

Example values:

  • 0: Danish, Finnish, Guarani, Hawaiian, Mandarin, Maori, Tlingit, VietnameseS, Wajarri, TokiPona, Greenlandic, Nama, Oneida
  • 0.1: Thai
  • 0.3: Spanish, Tagalog, Basque, Pirahã
  • 0.4: Irish, Korean
  • 0.6: Arabic, Archi, Ewe, Japanese, VietnameseN, Italian, Indonesian, Swahili, Malagasy
  • 0.8: !Xóõ
  • 0.9: Esperanto, French, Hindi, Russian, English, Portuguese, German, Turkish, Persian, Bengali, Telugu, Hausa, Georgian, Navajo
  • 1: Ubykh

Further reading:

Question 12: Aspiration

Aspiration is an audible burst of breath after a consonant release, especially in stops. This setting controls whether aspiration is contrastive and how much of the stop system participates in the contrast.

01

Example values:

  • 0: Arabic, Archi, Esperanto, Ewe, Finnish, French, Guarani, Hawaiian, Irish, Japanese, Maori, Russian, Spanish, Ubykh, Wajarri, TokiPona, Greenlandic, English, Portuguese, German, Italian, Indonesian, Tagalog, Hausa, Basque, Malagasy, Pirahã, Oneida, Turkish, Persian
  • 0.3: Danish, Mandarin, Korean, Thai, Swahili, Navajo, VietnameseN, VietnameseS, Georgian, Nama
  • 0.8: Hindi, Tlingit, Bengali, Telugu, !Xóõ

Further reading:

Question 13: Affricates

Affricates begin like stops and release into fricatives while functioning as single consonants. This setting controls whether sounds such as /ts/ and /tʃ/ are absent, marginal or part of a richer affricate inventory.

01

Example values:

  • 0: Arabic, Finnish, French, Guarani, Hawaiian, Irish, Japanese, Maori, Wajarri, TokiPona, Greenlandic, Portuguese, Pirahã, !Xóõ
  • 0.1: Danish, Russian, Spanish, VietnameseN, VietnameseS, Swahili, Oneida
  • 0.2: Hindi, English, Turkish, Persian, Indonesian
  • 0.4: Korean, Thai
  • 0.5: Archi, Esperanto, Ewe, Mandarin, Tlingit, Italian, Hausa, Malagasy, Navajo, Telugu, Tagalog
  • 0.6: Basque
  • 0.7: Ubykh, Bengali
  • 0.8: German
  • 0.9: Georgian
  • 1.0: Nama

Further reading:

Question 14: Liquids

Liquids include lateral sounds such as /l/ and rhotic sounds such as taps, trills or approximant r-sounds. This setting controls whether the language has a very small liquid system or several liquid contrasts.

01

Example values:

  • 0: Pirahã, Tlingit
  • 0.1: Guarani, German, Korean, Thai, Bengali, Telugu, Tagalog, Georgian, Navajo, Oneida
  • 0.2: Arabic, Danish, Esperanto, Finnish, French, Hawaiian, Hindi, Irish, Japanese, Mandarin, Maori, Russian, Ubykh, VietnameseN, TokiPona, Greenlandic, Portuguese, Italian, Turkish
  • 0.3: Archi, Ewe, VietnameseS, English, Persian, Indonesian, Hausa, Swahili, Nama, Malagasy, !Xóõ
  • 0.7: Wajarri
  • 0.8: Spanish, Basque

Further reading:

Question 15: Ejectives

Ejectives are voiceless consonants made with a glottalic airstream, often written with an apostrophe in phonetic transcription. This setting controls whether such consonants are available in the language.

01

Example values:

  • 0: Arabic, Danish, Esperanto, Ewe, Finnish, French, Guarani, Hawaiian, Hindi, Irish, Japanese, Mandarin, Maori, Russian, Spanish, VietnameseN, VietnameseS, Wajarri, TokiPona, Greenlandic, English, Portuguese, German, Korean, Italian, Turkish, Thai, Persian, Bengali, Telugu, Indonesian, Tagalog, Swahili, Nama, Basque, Malagasy, Pirahã, Oneida
  • 0.3: !Xóõ, Hausa
  • 0.6: Archi, Tlingit, Georgian, Navajo
  • 0.8: Ubykh

Further reading:

Question 16: Labialisation

Labialised consonants are pronounced with simultaneous lip rounding, as in sounds like /kʷ/. This setting controls whether labialisation is absent, marginal or a contrastive secondary articulation.

01

Example values:

  • 0: Arabic, Basque, Danish, English, Ewe, Finnish, French, German, Hawaiian, Hawaiian, Hindi, Irish, Italian, Japanese, Korean, Mandarin, Maori, Maori, Navajo, Pirahã, Russian, Spanish, Esperanto, Swahili, Turkish, VietnameseN, VietnameseS, Wajarri, TokiPona, Greenlandic, Portuguese, Indonesian, Bengali, Georgian, Malagasy, Nama, Telugu, Tagalog, Thai, Persian, Oneida, !Xóõ
  • 0.2: Tlingit, Guarani
  • 1: Ubykh, Archi
  • 0.4: Hausa

Further reading:

Question 17: Pharyngealisation

Pharyngealised consonants involve a secondary constriction in the pharynx, giving them a darker or more “emphatic” quality. This setting controls whether that contrast appears in the consonant system.

01

Example values:

  • 0: Basque, Danish, English, Ewe, Finnish, French, Georgian, German, Guarani, Hawaiian, Hawaiian, Hindi, Irish, Italian, Japanese, Korean, Mandarin, Maori, Maori, Navajo, Pirahã, Russian, Spanish, Esperanto, Swahili, Tlingit, Turkish, VietnameseN, VietnameseS, Wajarri, TokiPona, Greenlandic, Portuguese, Indonesian, Bengali, Malagasy, Nama, Telugu, Tagalog, Thai, Persian, Hausa, Oneida, !Xóõ
  • 0.2: Arabic
  • 1: Ubykh, Archi

Further reading:

Question 18: Implosives

Implosives are voiced stops produced with a glottalic ingressive component. This setting controls whether the language includes such sounds and, at higher values, whether they form a broader series.

01

Example values:

  • 0: Arabic, Archi, Danish, Esperanto, Ewe, Finnish, French, Guarani, Hawaiian, Hindi, Irish, Japanese, Mandarin, Maori, Russian, Spanish, Tlingit, Ubykh, Wajarri, TokiPona, Greenlandic, English, Portuguese, German, Korean, Italian, Turkish, Thai, Persian, Bengali, Telugu, Indonesian, Tagalog, Nama, Georgian, Basque, Malagasy, Pirahã, Navajo, !Xóõ, Oneida
  • 0.1: VietnameseN, VietnameseS
  • 0.6: Hausa, Swahili

Further reading:

Question 19: Clicks

Clicks are consonants made with a velaric ingressive airstream. They are rare globally but important in several southern African languages. This setting controls whether clicks appear in the inventory.

01

Example values:

  • 0: Arabic, Archi, Danish, Esperanto, Ewe, Finnish, French, Guarani, Hawaiian, Hindi, Irish, Japanese, Mandarin, Maori, Russian, Spanish, Tlingit, Ubykh, VietnameseN, VietnameseS, Wajarri, TokiPona, Greenlandic, English, Portuguese, German, Korean, Italian, Turkish, Thai, Persian, Bengali, Telugu, Indonesian, Tagalog, Hausa, Swahili, Georgian, Basque, Malagasy, Pirahã, Navajo, Oneida
  • 0.7: Nama, !Xóõ

Further reading:

Vowel systems

Vowels are produced with an open vocal tract and are organised by height, backness, rounding, length, nasality and reduction. Some languages have very small vowel systems, while others make many contrasts. This section controls the language’s vowel inventory and the way vowels behave in ordinary and weak syllables.

Vowel contrasts and behaviour

These settings define the vowel system: how many height and backness contrasts it has, whether rounding, length or nasality are contrastive, and whether vowels change in weak syllables or harmonise across a word.

Question 20: Vowel height

Vowel height describes how high the tongue is during a vowel. This setting controls whether the language has a simple two-height system, a common three-height system or a finer four-height distinction.

24

Example values:

  • 2: Arabic, Greenlandic, Wajarri, Pirahã
  • 3: Maori, Russian, Finnish, Hawaiian, Japanese, Spanish, TokiPona, Esperanto, Irish, Tlingit, !Xóõ, Basque, Hausa, Indonesian, Malagasy, Nama, Oneida, Persian, Swahili, Telugu, Turkish, Mandarin, Archi, Georgian, Guarani, Tagalog, Navajo, Ubykh
  • 4: French, Ewe, English, Danish, VietnameseN, VietnameseS, Hindi, Bengali, German, Italian, Korean, Portuguese, Thai

Further reading:

Question 21: Nasal vowels

Nasal vowels are produced with air flowing through the nose as well as the mouth. This setting controls whether nasal vowels are absent, limited to part of the system or parallel to the full oral vowel inventory.

01

Example values:

  • 0: Arabic, Greenlandic, Wajarri, Danish, Finnish, Maori, Russian, Hawaiian, Japanese, Spanish, Esperanto, Irish, Tlingit, VietnameseN, VietnameseS, TokiPona, English, Basque, Georgian, German, Hausa, Indonesian, Italian, Korean, Malagasy, Persian, Pirahã, Swahili, Tagalog, Telugu, Thai, Turkish, Mandarin, Archi, Ubykh
  • .4: French, Oneida, Portuguese
  • 1: Guarani, Ewe, !Xóõ, Bengali, Navajo, Hindi
  • 0.6: Nama

Further reading:

Question 22: Vowel length

Vowel length distinguishes short and long vowels. Low values ignore length as a contrast; high values make vowel quantity a regular part of the phonology and word shapes.

01

Example values:

  • 0.0: Russian, Ubykh, Spanish, Esperanto, Guarani, Mandarin, French, TokiPona, English, Hindi, Bengali, Georgian, Indonesian, Italian, Malagasy, Persian, Pirahã, Portuguese, Swahili, Wajarri, Basque
  • 0.4: Danish, German, Korean, Oneida, Turkish, VietnameseN, VietnameseS, Ewe
  • 1: Arabic, Greenlandic, Maori, Hawaiian, Finnish, Japanese, Irish, Tlingit, Archi, Nama, Telugu, Tagalog, Thai, Hausa, Navajo, !Xóõ

Further reading:

Question 23: Vowel backness

Backness describes whether vowels are articulated towards the front, centre or back of the mouth. This setting controls whether the language has a vertical vowel system, a front/back contrast or a three-way front/central/back distinction.

13

Example values:

  • 1: Ubykh
  • 2: Arabic, Greenlandic, Wajarri, Maori, Russian, Hawaiian, Finnish, Japanese, Spanish, Esperanto, Irish, Tlingit, French, Danish, Ewe, VietnameseS, TokiPona, English, Hindi, !Xóõ, Basque, Bengali, German, Hausa, Italian, Korean, Malagasy, Nama, Navajo, Oneida, Persian, Pirahã, Portuguese, Swahili, Tagalog, Telugu, Thai, Turkish, Georgian
  • 3: Guarani, VietnameseN, Mandarin, Indonesian, Archi

Further reading:

Question 24: Vowel-system shape

This setting controls the overall geometry of the vowel system. Triangular systems crowd contrasts towards the high vowels and have few low vowels; square systems distribute vowels more evenly across frontness and height.

13

Example values:

  • 1: Pirahã
  • 2: Maori, Russian, Arabic, Greenlandic, Wajarri, Hawaiian, Japanese, Spanish, Esperanto, Irish, Tlingit, Ewe, French, TokiPona, Hindi, !Xóõ, Basque, Bengali, German, Hausa, Indonesian, Italian, Korean, Malagasy, Nama, Navajo, Oneida, Portuguese, Swahili, Tagalog, Thai, Turkish, Mandarin, Archi, Georgian
  • 3: Guarani, Danish, Ubykh, Finnish, English, VietnameseN, VietnameseS, Telugu, Persian

Further reading:

Question 25: Reduced vowels

Many languages reduce vowels in unstressed or weak syllables. Low values allow only a restricted reduced-vowel set; high values keep the full vowel inventory available even in weak positions.

01

Example values:

  • .2: Danish, Portuguese, Irish
  • .4: English, Russian, German, Mandarin, Malagasy
  • .6: Italian
  • 1: French, Hindi, !Xóõ, Basque, Bengali, Georgian, Hausa, Korean, Nama, Navajo, Oneida, Persian, Pirahã, Swahili, Tagalog, Telugu, Thai, Turkish, Indonesian
  • 1: Guarani, Ewe, Ubykh, Archi, Maori, Finnish, Hawaiian, Japanese, Spanish, TokiPona, Esperanto, Tlingit, Arabic, Greenlandic, Wajarri, VietnameseN, VietnameseS

Further reading:

Question 26: Empty nuclei

Empty nuclei are weak underlying syllable centres that may not surface as audible vowels. Low values require ordinary vowels; higher values allow more vowel deletion and, at the extreme, words with very little overt vocalic material.

01

Example values:

  • 0.0: Arabic, Greenlandic, Archi, Basque, Danish, English, Ewe, Finnish, French, German, Guarani, Hawaiian, Hawaiian, Hindi, Irish, Italian, Korean, Mandarin, Maori, Maori, Navajo, Pirahã, Russian, Spanish, Esperanto, Swahili, Tlingit, Turkish, Ubykh, VietnameseN, VietnameseS, Wajarri, French, English, TokiPona, Portuguese, Indonesian, Bengali, Georgian, Malagasy, Nama, Telugu, Tagalog, Thai, Persian, Hausa, Oneida, !Xóõ
  • 0.2: Japanese

Further reading:

Question 27: Roundedness

Roundedness concerns whether lip rounding distinguishes vowels beyond ordinary front/back and height contrasts. The language may contrast front rounded vowels, back unrounded vowels or avoid roundedness as an independent contrast.

the language distinguishes between front rounded and unrounded vowels (/i/ vs. /y/)– e.g., Danish, Finnish, French, German, Turkish, Mandarin
the language distinguishes between back rounded and unrounded vowels (/u/ vs. /ɯ/)– e.g., Thai, VietnameseN, VietnameseS, Korean
the language should not use roundedness– e.g., Ewe, Maori, Hawaiian, Arabic, Greenlandic, Wajarri, Ubykh, Archi, Japanese, Spanish, Esperanto, Irish, Tlingit, Russian, Guarani, TokiPona, English, Hindi, !Xóõ, Basque, Bengali, Georgian, Hausa, Indonesian, Italian, Malagasy, Nama, Navajo, Persian, Pirahã, Portuguese, Swahili, Tagalog, Telugu, Oneida

Further reading:

Question 28: Vowel harmony

Vowel harmony requires vowels in a word to agree in features such as frontness, rounding or tongue-root position. This setting controls whether vowels are selected independently or constrained by harmony patterns.

No– e.g., Danish, Maori, Russian, Hawaiian, Japanese, Ubykh, Archi, Spanish, Esperanto, Irish, Tlingit, Arabic, Greenlandic, Wajarri, French, Ewe, VietnameseN, VietnameseS, English, TokiPona, Hindi, German, Mandarin, Portuguese, Indonesian, Basque, Italian, Korean, Swahili, Bengali, Georgian, Malagasy, Nama, Pirahã, Tagalog, Thai, Persian, Hausa, Navajo, Oneida
Yes– e.g., Guarani, Finnish, Turkish, Telugu, !Xóõ

Further reading:

Syllable structure and prosody

Beyond individual consonants and vowels, languages differ in how sounds combine into syllables and how words use stress, pitch and phonation. These settings influence the rhythm of the language, the complexity of possible word shapes and the extent to which pitch or glottalisation can distinguish meanings.

Syllable complexity

Syllable structure determines which combinations of vowels and consonants are possible. Some languages strongly prefer simple CV syllables; others allow diphthongs and consonants after the vowel, including complex codas.

Question 29: Diphthongs

Diphthongs are vowel sequences that form one syllable, such as a vowel moving towards /i/ or /u/. This setting controls whether the language allows such vowel-glide nuclei or prefers simple monophthongs.

01

Example values:

  • 0: Greenlandic, Wajarri, Danish, Russian, Japanese, Esperanto, Tlingit, TokiPona, French, Hindi, Georgian, Oneida, Persian, Swahili, Turkish, Ubykh
  • 0.2: Hawaiian, Arabic, Finnish, Spanish, Mandarin, VietnameseN, VietnameseS, Basque, Irish, Archi, Ewe, Malagasy, Telugu, Hausa
  • .4: English, German, Indonesian, Maori, Portuguese, Italian, Korean, Bengali, Guarani, Nama, Pirahã, Tagalog, Thai, Navajo, !Xóõ
Question 30: Coda complexity

A coda is the consonantal part of a syllable after the vowel. Low values give the language open syllables; higher values allow final consonants and increasingly complex syllable endings.

01

Example values:

  • 0.0: Hawaiian, Maori, Guarani, Malagasy, Pirahã
  • 0.1: Japanese, TokiPona
  • 0.3: Swahili, Italian, Wajarri, Greenlandic, Spanish, Portuguese, Basque
  • 0.4: Esperanto, Arabic, Finnish, Hindi, Turkish, Indonesian, Archi, Bengali, Nama, Telugu, Tagalog, Persian, Navajo
  • 0.6: English, French, Irish, Tlingit, Ubykh
  • 0.8: Russian, Danish, Oneida
  • 1.0: German, Georgian
  • 0.2: Mandarin, VietnameseN, VietnameseS, Korean, Ewe, Thai, Hausa, !Xóõ

Further reading:

Question 31: Sibilant + other clusters

Many languages allow sibilant-initial clusters, especially /s/ + stop or /s/ + nasal. Others forbid them. This slider controls the range of sibilant+X onsets.

01

Example values:

  • 0.0: Hawaiian, Turkish, Japanese, Maori, TokiPona, Arabic, Wajarri, Greenlandic, Finnish, Mandarin, VietnameseN, VietnameseS, Korean, Archi, Ewe, Guarani, Malagasy, Nama, Pirahã, Thai, Persian, Hausa, !Xóõ
  • 0.3: Spanish, Portuguese, Basque
  • 0.5: French, Esperanto, Hindi, Indonesian, Irish, Italian, Swahili, Bengali, Telugu, Tlingit, Tagalog, Navajo
  • 0.6: English, Russian, Danish, German
  • 1.0: Georgian, Oneida, Ubykh

Stress, tone and phonation

Prosodic systems use prominence, pitch or voice quality to distinguish words or organise speech. Stress makes syllables prominent; tone uses pitch contrastively; phonation contrasts use features such as glottalisation or creaky voice.

Question 32: Stress

Stress makes one syllable more prominent than others. This setting controls whether the generated language uses stress or accent as part of its word-level phonology.

01

Example values:

  • 0: French, Greenlandic, Japanese, VietnameseN, VietnameseS, Hindi, Mandarin, Indonesian, Korean, Ewe, Nama, Telugu, Hausa, !Xóõ
  • 1: English, Danish, Esperanto, Hawaiian, Maori, TokiPona, Arabic, Wajarri, Finnish, Russian, Spanish, German, Turkish, Portuguese, Basque, Irish, Italian, Swahili, Archi, Bengali, Georgian, Guarani, Malagasy, Pirahã, Tlingit, Tagalog, Thai, Persian, Navajo, Oneida, Ubykh
Question 33: Tone

Tone uses pitch differences to distinguish word meanings. This setting controls whether the language is non-tonal or uses a small, medium or rich set of lexical tone contrasts.

06

Example values:

  • 0: French, Arabic, Greenlandic, English, Wajarri, Finnish, Maori, Hawaiian, Japanese, Esperanto, TokiPona, Russian, Spanish, Hindi, German, Turkish, Portuguese, Indonesian, Basque, Irish, Italian, Korean, Swahili, Archi, Bengali, Georgian, Guarani, Malagasy, Telugu, Tagalog, Persian, Oneida, Ubykh
  • 2: Danish, Ewe, Pirahã, Tlingit, Navajo
  • 4: Mandarin, Hausa, !Xóõ
  • 6: VietnameseN, VietnameseS, Nama, Thai
Question 34: Sandhi

To what extent should morphemes blend together? For instance, in Greenlandic, Nuuk and -mi “in” combine to produce Nuummi “in Nuuk”.

01

Example values:

  • 0: Maori, Hawaiian, English, Esperanto, Japanese, Danish, TokiPona, Arabic, Wajarri, Finnish, Russian, Spanish, Hindi, German, Turkish, Mandarin, Portuguese, Indonesian, VietnameseN, VietnameseS, Basque, Irish, Italian, Korean, Swahili, Archi, Bengali, Ewe, Georgian, Guarani, Malagasy, Nama, Pirahã, Telugu, Tlingit, Tagalog, Thai, Persian, Hausa, Navajo, Oneida, !Xóõ, Ubykh
  • 0.8: French, Greenlandic
Question 35: Phonation contrast

This setting adds voice-quality contrasts such as glottalisation or creaky phonation. If the language has tones, phonation can distinguish tonal categories; otherwise it behaves more like a vowel-level contrast similar to length or stød.

No– e.g., English, Mandarin, Russian, Hawaiian, Spanish, French, TokiPona, Greenlandic, Japanese, Esperanto, Maori, Arabic, Wajarri, Finnish, Hindi, German, Turkish, Portuguese, Indonesian, VietnameseS, Basque, Irish, Italian, Korean, Swahili, Archi, Bengali, Ewe, Georgian, Guarani, Malagasy, Nama, Pirahã, Telugu, Tlingit, Tagalog, Thai, Persian, Hausa, Navajo, Oneida, Ubykh
Yes– e.g., Danish, VietnameseN, !Xóõ

Morphology

Morphological structure

Morphology concerns how words are built from smaller meaningful parts. This section controls where grammatical markers appear, how tightly they attach, how much information they express and how densely words can be packed with affixes or incorporated material.

Marker position, clitics and obligatoriness

Bound grammatical material can appear before or after its host, attach tightly as affixes or loosely as clitics, and be obligatory or optional. These choices influence the overall morphological profile of the language.

Question 36: Marker position

This setting controls where grammatical markers tend to appear relative to their host. Low values favour suffixes and enclitics; high values favour prefixes and proclitics; middle values allow a more mixed system.

01

Example values:

  • 0: Japanese, Greenlandic, Wajarri, Finnish, Hindi, Turkish, Mandarin, Korean, Bengali, Pirahã, Telugu, !Xóõ
  • 0.2: English, Danish, Russian, Spanish, Portuguese, Basque, Italian, Archi, Nama, Hausa
  • 0.4: French, Esperanto, German, Irish, Ewe, Georgian, Guarani, Thai, Persian
  • 0.8: Maori, Hawaiian, Arabic, TokiPona, Indonesian, VietnameseN, VietnameseS, Swahili, Tlingit, Oneida
  • 0.6: Malagasy, Tagalog, Ubykh
  • 1: Navajo
Question 37: Clitics

Clitics are phonologically dependent elements that behave more like small words than ordinary affixes. This setting controls whether grammatical material is mostly tightly affixed or often expressed through clitics.

01

Example values:

  • 0.4: Esperanto, Danish, Arabic, Wajarri, Spanish, Hindi, Portuguese, Indonesian, Irish, Italian, Korean, Bengali, Malagasy, Telugu, Tlingit, Navajo, Oneida, !Xóõ
  • 0.6: French, English, Ewe, Guarani, Thai, Persian, Hausa
  • 0.8: Maori, Hawaiian, Mandarin, VietnameseN, VietnameseS, Nama, Tagalog
  • 1.0: TokiPona, Japanese
  • 0.2: Greenlandic, Finnish, Russian, German, Turkish, Basque, Swahili, Archi, Georgian, Pirahã, Ubykh
Question 38: Marking obligatoriness

Some languages require grammatical categories to be overtly marked, while others leave common or recoverable values unmarked. This setting controls how often morphology must appear, especially for default or contextually obvious meanings.

01

Example values:

  • 1.0: Maori, Hawaiian, TokiPona, Mandarin, Indonesian, VietnameseN, VietnameseS
  • 0.6: French, Esperanto, English, Japanese, Korean, Bengali, Ewe, Pirahã
  • 0.2: Danish, Arabic, Wajarri, Greenlandic, Finnish, Russian, Spanish, Hindi, German, Turkish, Portuguese, Basque, Irish, Italian, Swahili, Archi, Georgian, Malagasy, Nama, Telugu, Tlingit, Tagalog, Persian, Hausa, Navajo, Oneida, !Xóõ, Ubykh
  • 0.4: Guarani
  • 0.8: Thai

Fusion

Fusion describes how much grammatical information is bundled into each marker. In transparent agglutinative systems, each affix tends to express one function; in more fusional systems, a single ending may express several categories at once.

Question 39: Fusion

Fusion describes how much information a single marker expresses. Low values favour transparent, agglutinative morphology; higher values allow endings that combine categories such as case, number, person or tense.

01

Example values:

  • 0: Maori, Hawaiian, Esperanto, TokiPona, Wajarri, Turkish, Mandarin, Indonesian, VietnameseN, VietnameseS, Thai
  • 0.2: Japanese, Finnish, Korean, Swahili, Archi, Ewe, Guarani, Pirahã, !Xóõ, Ubykh
  • 0.4: French, English, Danish, Greenlandic, Hindi, Basque, Bengali, Malagasy, Telugu, Tagalog, Persian
  • 0.8: Arabic, Russian, Spanish, German, Portuguese, Irish, Italian, Nama
  • 0.6: Georgian, Tlingit, Hausa, Navajo, Oneida

Affix density and incorporation

Languages can make words complex in different ways. One route is to add many affixes to a root; another is to incorporate nouns or arguments into the verb. Both can create long words, but they represent different grammatical strategies.

Question 40: Affix density

Affix density controls how much grammatical and derivational material can be packed into individual words. High values create words with many ordered prefixes or suffixes; low values keep words relatively bare.

01

Example values:

  • 0.0: TokiPona, English, Mandarin, VietnameseN, VietnameseS, Thai
  • 0.2: Maori, Hawaiian, Japanese, Danish, Arabic, Wajarri, French, Spanish, Hindi, Portuguese, Irish, Italian, Ewe
  • 0.4: Esperanto, Finnish, Russian, German, Indonesian, Bengali, Nama, Persian, Hausa, !Xóõ
  • 1.0: Greenlandic, Tlingit, Navajo, Oneida, Ubykh
  • 0.6: Turkish, Basque, Korean, Swahili, Archi, Guarani, Malagasy, Pirahã, Telugu, Tagalog
  • 0.8: Georgian
Question 41: Incorporation

Incorporation lets a verb absorb a noun or argument into the verbal word. This differs from merely adding many affixes: it changes how participants are expressed in the structure of the verb itself.

01

Example values:

  • 0: Japanese, Maori, Hawaiian, Arabic, Wajarri, Danish, Esperanto, French, Greenlandic, English, TokiPona, Finnish, Russian, Spanish, Hindi, German, Turkish, Mandarin, Portuguese, Indonesian, VietnameseN, VietnameseS, Basque, Irish, Italian, Korean, Swahili, Archi, Bengali, Ewe, Georgian, Malagasy, Nama, Pirahã, Telugu, Tagalog, Thai, Persian, Hausa, Navajo, !Xóõ
  • 0.2: Guarani, Ubykh
  • 0.4: Tlingit
  • 0.6: Oneida

Grammatical categories

Languages vary in which grammatical categories they require speakers to mark. Number, gender, definiteness and property words may be central parts of the grammar, marginal features or absent altogether. These settings determine which categories appear in noun phrases and how prominently they shape the generated grammar.

Number

Number marking shows whether nouns, pronouns or agreement forms distinguish one, two or more participants or objects.

Question 42: Number

Number marking distinguishes how many entities are involved. The language may ignore grammatical number, distinguish singular and plural or add a dual category for exactly two.

it should not have number as a grammatical category– e.g., Japanese, TokiPona, Pirahã
the language distinguishes between singular and plural– e.g., Esperanto, English, Basque, French, English, Danish, Greenlandic, Finnish, Russian, Spanish, Hindi, German, Turkish, Mandarin, Portuguese, Indonesian, VietnameseN, VietnameseS, Irish, Italian, Korean, Swahili, Archi, Bengali, Ewe, Georgian, Guarani, Malagasy, Telugu, Tlingit, Tagalog, Thai, Persian, Hausa, !Xóõ, Ubykh
the language employs singular, dual and plural– e.g., Hawaiian, Maori, Arabic, Wajarri, Nama, Navajo, Oneida

Gender and noun classes

Gender systems classify nouns into agreement classes. Some systems are based on sex or animacy; others divide nouns into larger sets of grammatical classes with less obvious semantic motivation.

Question 43: Gender

Gender and noun-class systems sort nouns into grammatical categories that may control agreement. The system may be absent, based on masculine/feminine or animate/inanimate distinctions, or expanded into several genders or many noun classes.

it should not have gender as a grammatical category– e.g., Finnish, Maori, Hawaiian, Esperanto, Japanese, TokiPona, English, Wajarri, Greenlandic, Turkish, Mandarin, Indonesian, VietnameseN, VietnameseS, Basque, Korean, Bengali, Ewe, Georgian, Guarani, Malagasy, Pirahã, Tlingit, Tagalog, Thai, Persian, Navajo, Ubykh
masc/fem– e.g., French, Spanish, Irish, Arabic, Hindi, Portuguese, Italian, Hausa
anim/inanim– e.g., Danish
the language employs masculine, feminine and neuter– e.g., Russian, German, Nama, Telugu
the language has four genders– e.g., Archi, Oneida
the language has noun classes– e.g., Swahili, !Xóõ

Articles and definiteness

Articles and definiteness markers indicate whether a noun phrase refers to something identifiable, specific or newly introduced. Languages may mark definiteness, indefiniteness, both or neither.

Question 44: Definite articles

Definite marking signals that a noun phrase refers to something identifiable or already established. This setting controls whether the language marks definiteness with an article, affix or similar marker.

Yes– e.g., Arabic, Maori, Esperanto, French, Spanish, Hawaiian, English, Danish, English, German, Portuguese, Indonesian, Basque, Irish, Italian, Bengali, Ewe, Guarani, Malagasy, Nama, Tlingit, Hausa, Ubykh
No– e.g., Russian, Finnish, TokiPona, Mandarin, Japanese, Wajarri, Greenlandic, Hindi, Turkish, VietnameseN, VietnameseS, Korean, Swahili, Archi, Georgian, Pirahã, Telugu, Tagalog, Thai, Persian, Navajo, Oneida, !Xóõ

Further reading:

Question 45: Indefinite articles

Indefinite marking introduces or marks non-specific nouns, often like English “a” or “an”. This setting controls whether the language has a dedicated indefinite marker.

Yes– e.g., Spanish, English, Danish, English, French, Hawaiian, Maori, German, Turkish, Portuguese, Basque, Italian, Ewe, Persian, Hausa, Ubykh
No– e.g., Arabic, Wajarri, Finnish, TokiPona, Japanese, Esperanto, Russian, Mandarin, Hindi, Greenlandic, Indonesian, VietnameseN, VietnameseS, Irish, Korean, Swahili, Archi, Bengali, Georgian, Guarani, Malagasy, Nama, Pirahã, Telugu, Tlingit, Tagalog, Thai, Navajo, Oneida, !Xóõ

Further reading:

Property words

Languages do not all treat property concepts such as “big”, “red” or “good” as a separate adjective class. They may behave more like verbs, more like nouns or as their own category.

Question 46: Adjectives

Languages differ in how they express property concepts such as size, colour and quality. These meanings may behave like verbs, like nouns or as a separate adjective class with its own grammar.

Adjectives are verbs– e.g., Maori, Hawaiian, Mandarin, Japanese, TokiPona, Greenlandic, VietnameseN, VietnameseS, Korean, Guarani, Thai, Navajo, Oneida
Adjectives are nouns
Adjectives are separate from both– e.g., Esperanto, English, French, English, Danish, Arabic, Wajarri, Finnish, Russian, Spanish, Hindi, German, Turkish, Portuguese, Indonesian, Basque, Irish, Italian, Swahili, Archi, Bengali, Ewe, Georgian, Malagasy, Nama, Pirahã, Telugu, Tlingit, Tagalog, Persian, Hausa, !Xóõ, Ubykh

Alignment and marking

Clause alignment

Alignment describes how languages treat the participants of a clause: the subject of an intransitive verb, the agent of a transitive verb and the object of a transitive verb. These settings control whether noun phrases and, eventually, verbal marking follow neutral, accusative, ergative, active or split patterns.

Noun-phrase alignment

Noun-phrase alignment concerns how the core participants of a clause are marked on the noun phrases themselves. It asks whether intransitive subjects pattern with transitive subjects, with objects or according to agency and aspect.

Question 47: Noun-phrase alignment

Alignment controls how noun phrases mark core clause roles. Accusative systems group intransitive subjects with transitive agents; ergative systems group intransitive subjects with objects; active and split systems divide the pattern further.

noun phrases are not marked for grammatical roles (neutral alignment)– e.g., French, Mandarin, Indonesian, VietnameseN, VietnameseS, Swahili, Pirahã, Thai, Navajo, Oneida, !Xóõ
nominative–accusative: subjects are treated alike, objects are marked– e.g., Danish, Esperanto, Finnish, Japanese, English, Hawaiian, TokiPona, Arabic, Russian, Spanish, German, Turkish, Portuguese, Irish, Italian, Korean, Bengali, Ewe, Guarani, Malagasy, Nama, Telugu, Persian, Hausa
ergative–absolutive: objects and intransitive subjects are treated alike, agents are marked– e.g., Maori, Wajarri, Greenlandic, Basque, Archi, Tlingit, Tagalog, Ubykh
active–stative: intransitive subjects are marked differently depending on agency or volition
split alignment: ergative alignment applies in the past/perfective tense only– e.g., Hindi, Georgian

Possession

Possessive constructions express relations such as kinship, ownership, part-whole structure and association. Languages may mark the possessor, the possessed noun, both or neither. These settings determine where possession is shown inside noun phrases.

Possessor marking

This setting concerns marking placed on the possessor: the owner, kin relation, whole or associated noun phrase.

Question 48: Possessor marking

This setting controls whether the possessor itself receives marking in possessive noun phrases, as with a genitive case, possessive particle or similar dependent marking.

it should mark the possessor– e.g., Danish, Japanese, Arabic, French, Wajarri, Esperanto, English, Hawaiian, Maori, Greenlandic, Finnish, Russian, Spanish, Hindi, German, Turkish, Mandarin, Portuguese, VietnameseN, VietnameseS, Basque, Irish, Italian, Korean, Swahili, Archi, Bengali, Ewe, Georgian, Malagasy, Nama, Telugu, Tagalog, Thai, !Xóõ, Ubykh
it should not mark the possessor– e.g., TokiPona, Indonesian, Guarani, Pirahã, Tlingit, Persian, Hausa, Navajo, Oneida

Possessed-noun marking

This setting concerns marking placed on the possessed noun: the thing owned, related or possessed.

Question 49: Possessed-noun marking

This setting controls whether the possessed noun carries marking that shows it is possessed. Such marking may appear as a suffix, prefix or special possessed form of the noun.

the language marks the possessed noun ("the man his house")– e.g., Arabic, Greenlandic, Finnish, Turkish, Indonesian, Guarani, Malagasy, Tlingit, Persian, Hausa, Navajo, Oneida, Ubykh
it should not mark the possessed noun– e.g., TokiPona, Danish, Japanese, French, Esperanto, English, Hawaiian, Maori, Wajarri, Russian, Spanish, Hindi, German, Mandarin, Portuguese, VietnameseN, VietnameseS, Basque, Irish, Italian, Korean, Swahili, Archi, Bengali, Ewe, Georgian, Nama, Pirahã, Telugu, Tagalog, Thai, !Xóõ

Adpositions and adpositional phrases

Adpositions include prepositions, postpositions and similar relational words. They express meanings such as location, direction, source, instrument and accompaniment. Languages differ in whether adpositions come before or after the noun phrase, whether they govern case and whether complex meanings are built compositionally.

Marking in adpositional phrases

Adpositional phrases can mark the relationship on the adposition itself, on the noun phrase governed by the adposition or on neither. These are head-marking and dependent-marking strategies.

Question 50: Head-marking on adpositions

Head-marking in adpositional phrases means the adposition itself shows information about its complement. This can produce different adpositional forms depending on person, number or pronoun-like reference.

yes– e.g., Arabic, Greenlandic, Finnish, Irish, Guarani, Malagasy, Persian, Hausa, Navajo, !Xóõ, Ubykh
no– e.g., TokiPona, Danish, French, Japanese, Turkish, Esperanto, English, Hawaiian, Maori, Wajarri, Russian, Spanish, Hindi, German, Mandarin, Portuguese, Indonesian, VietnameseN, VietnameseS, Basque, Italian, Korean, Swahili, Archi, Bengali, Ewe, Georgian, Nama, Pirahã, Telugu, Tlingit, Tagalog, Thai, Oneida
Question 51: Dependent-marking on adpositions

Dependent-marking in adpositional phrases means the noun phrase governed by the adposition receives case or other marking. This setting controls whether adpositions impose such marking on their complements.

yes– e.g., German, Arabic, Wajarri, Greenlandic, Finnish, Russian, Hindi, Turkish, Basque, Irish, Archi, Bengali, Georgian, Malagasy, Nama, Telugu, Tlingit, Tagalog, Hausa, Ubykh
no– e.g., TokiPona, English, Mandarin, Japanese, French, Esperanto, Hawaiian, Danish, Maori, Spanish, Portuguese, Indonesian, VietnameseN, VietnameseS, Italian, Korean, Swahili, Ewe, Guarani, Pirahã, Thai, Persian, Navajo, Oneida, !Xóõ

Prepositions and postpositions

Adpositions may come before the noun phrase, as prepositions do, or after it, as postpositions do. This choice often correlates with broader word-order tendencies.

Question 52: Adposition position

This setting controls whether relational words come before or after the noun phrase. Prepositions precede their complement; postpositions follow it. The choice often fits the language’s wider word-order profile.

It should use prepositions– e.g., English, TokiPona, Danish, French, Arabic, Esperanto, Spanish, Hawaiian, Maori, Russian, German, Mandarin, Portuguese, Indonesian, VietnameseN, VietnameseS, Irish, Italian, Swahili, Malagasy, Tagalog, Thai, Persian, Hausa, !Xóõ
It should use postpositions– e.g., Japanese, Korean, Hindi, Wajarri, Greenlandic, Finnish, Turkish, Basque, Archi, Bengali, Ewe, Georgian, Guarani, Nama, Pirahã, Telugu, Tlingit, Navajo, Oneida, Ubykh

Complex adpositional meanings

Complex spatial or relational meanings can be expressed by separate lexical adpositions or built compositionally from simpler pieces.

Question 53: Complex adpositions

Some languages build complex relational meanings from simpler elements, such as “from inside” rather than a separate word “out of”. This setting controls whether complex adpositional meanings are decomposed.

yes– e.g., Japanese, TokiPona, Hawaiian, Maori, Arabic, Greenlandic, Finnish, Hindi, Turkish, Mandarin, Indonesian, VietnameseN, VietnameseS, Basque, Irish, Korean, Swahili, Archi, Bengali, Ewe, Georgian, Guarani, Malagasy, Nama, Pirahã, Telugu, Tlingit, Tagalog, Thai, Persian, Hausa, Navajo, Ubykh
no– e.g., English, German, Esperanto, English, French, Danish, Wajarri, Russian, Spanish, Portuguese, Italian, Oneida, !Xóõ

Syntax and word order

Argument expression

Languages differ in how much information must be stated overtly. In some languages pronouns are normally expressed, while in others recoverable subjects or objects can be left out. This section controls how freely the generated language may omit pronouns when context or verbal marking makes the reference clear.

Pro-drop

Pro-drop, or zero anaphora, allows recoverable pronouns to be omitted. The result is a grammar where context, agreement or discourse structure can carry more of the reference.

Question 54: Pro-drop

Pro-drop allows pronouns to be omitted when they are recoverable from context, agreement or discourse. Low values require overt pronouns more often; high values permit more zero subjects or objects.

01

Example values:

  • 0: TokiPona, English, French, Esperanto, Danish, German, Ewe, Malagasy, Nama, Hausa, !Xóõ
  • 0.2: Arabic, Finnish, Russian, Spanish, Portuguese, Irish, Italian, Persian
  • 1: Japanese, Hawaiian, Maori, Wajarri, Greenlandic, Hindi, Turkish, Mandarin, Indonesian, VietnameseN, VietnameseS, Basque, Korean, Swahili, Archi, Bengali, Georgian, Guarani, Pirahã, Telugu, Tlingit, Tagalog, Thai, Navajo, Oneida, Ubykh

Basic word order

Word order sets the default arrangement of subjects, verbs, objects and noun-phrase modifiers. These choices do not make every sentence rigidly identical, but they establish the ordinary patterns used by the generated grammar and examples.

Subject, verb and object

The basic order of subject, verb and object sets the ordinary shape of transitive clauses. Other orders may still occur in special contexts, but this is the default used by the generator.

Question 55: Basic constituent order

This setting chooses the default order of subject, verb and object in transitive clauses. It establishes the ordinary clause pattern used by the generator, even if other orders could occur pragmatically.

it should use OSV
it should use OVS
it should use SOV– e.g., Japanese, Wajarri, Greenlandic, Hindi, Turkish, Basque, Korean, Archi, Bengali, Georgian, Nama, Pirahã, Telugu, Tlingit, Persian, Navajo, Ubykh
it should use SVO– e.g., Danish, French, Finnish, Esperanto, TokiPona, English, Russian, Spanish, German, Mandarin, Portuguese, Indonesian, VietnameseN, VietnameseS, Italian, Swahili, Ewe, Guarani, Thai, Hausa, !Xóõ
it should use VOS– e.g., Malagasy
it should use VSO– e.g., Maori, Hawaiian, Arabic, Irish, Tagalog, Oneida

Demonstrative order

Demonstratives are words such as “this” and “that”. This setting controls whether they normally appear before or after the noun they modify.

Question 56: Demonstrative order

Demonstratives are words such as “this” and “that”. This setting controls whether demonstratives normally stand before the noun they modify or follow it inside the noun phrase.

it should place demonstratives before the noun– e.g., Danish, French, Esperanto, Japanese, English, Hawaiian, Maori, Arabic, Finnish, Russian, Spanish, Hindi, German, Turkish, Mandarin, Portuguese, Italian, Korean, Archi, Bengali, Georgian, Guarani, Malagasy, Nama, Telugu, Tlingit, Tagalog, Persian, Hausa, Navajo, Oneida, Ubykh
it should place demonstratives after the noun– e.g., TokiPona, Wajarri, Greenlandic, Indonesian, VietnameseN, VietnameseS, Basque, Irish, Swahili, Ewe, Pirahã, Thai, !Xóõ

Adjective order

If the language has adjective-like modifiers, this setting controls whether they normally appear before or after the noun.

Question 57: Adjective order

This setting controls the ordinary position of adjective-like modifiers. If the language has a separate adjective class, adjectives may normally precede the noun or follow it.

Adjective + Noun– e.g., Danish, Esperanto, Japanese, English, Finnish, Russian, Hindi, German, Turkish, Mandarin, Korean, Archi, Bengali, Georgian, Nama, Telugu, Tagalog, Hausa
Noun + Adjective– e.g., French, Hawaiian, Maori, TokiPona, Arabic, Wajarri, Greenlandic, Spanish, Portuguese, Indonesian, VietnameseN, VietnameseS, Basque, Irish, Italian, Swahili, Ewe, Guarani, Malagasy, Pirahã, Tlingit, Thai, Persian, Navajo, Oneida, !Xóõ, Ubykh

Genitive order

Genitive order controls whether possessors and similar noun-phrase dependents normally come before or after the possessed noun.

Question 58: Genitive order

Genitives express possession and related noun-noun relationships. This setting controls whether the possessor normally comes before the possessed noun or after it.

it should place the possessor after the possessed– e.g., TokiPona, French, Arabic, Esperanto, Hawaiian, Maori, Russian, Spanish, German, Portuguese, Indonesian, VietnameseN, VietnameseS, Irish, Italian, Swahili, Malagasy, Tagalog, Thai, Persian, Hausa
it should place the possessor before the possessed– e.g., Japanese, Danish, English, Wajarri, Greenlandic, Finnish, Hindi, Turkish, Mandarin, Basque, Korean, Archi, Bengali, Ewe, Georgian, Guarani, Nama, Pirahã, Telugu, Tlingit, Navajo, Oneida, !Xóõ, Ubykh

Numeral order

Numeral order controls whether number words normally appear before or after the noun they count.

Question 59: Numeral order

This setting controls where numerals appear in the noun phrase. Number words may normally precede the noun they count or follow it.

Numeral + Noun– e.g., Danish, French, Esperanto, English, Hawaiian, Arabic, Finnish, Russian, Spanish, Hindi, German, Turkish, Mandarin, Portuguese, Indonesian, VietnameseN, VietnameseS, Basque, Irish, Italian, Archi, Bengali, Georgian, Guarani, Nama, Pirahã, Telugu, Tlingit, Tagalog, Persian, Navajo, Oneida, Ubykh
Noun + Numeral– e.g., Japanese, Maori, TokiPona, Wajarri, Greenlandic, Korean, Swahili, Ewe, Malagasy, Thai, Hausa, !Xóõ

Relative, complement and quotative clauses

These settings control how the language builds clauses inside other clauses: relative clauses modifying nouns, complement clauses functioning as arguments and quotative constructions marking reported speech.

Question 60: Relative clause strategy

Relative clauses modify nouns, as in “the person who came”. Languages may place the relative clause after the noun, before it, inside a head-internal construction or in a correlative structure paired with a demonstrative.

Postnominal relative clause (after the noun, often with a relative pronoun) – e.g., Finnish, Danish, French, English, Esperanto, Hawaiian, Maori, Arabic, Wajarri, Greenlandic, Russian, Spanish, German, Portuguese, Indonesian, VietnameseN, VietnameseS, Irish, Italian, Swahili, Ewe, Georgian, Guarani, Malagasy, Tagalog, Thai, Persian, Hausa, !Xóõ
Prenominal relative clause (before the noun, no relative pronoun) – e.g., Japanese, Basque, Turkish, Mandarin, Korean, Archi, Nama, Telugu, Tlingit, Ubykh
Embedded/head-internal strategy.– e.g., Navajo, Oneida
Correlative strategy (a relative clause introduced by a relativiser, paired with a demonstrative in the main clause) – e.g., Hindi, TokiPona, Bengali, Pirahã

Further reading:

Question 61: Complement clause strategy

Complement clauses function as arguments of verbs, as in “I know that he came”. This setting chooses whether such clauses use a complementiser, a nominalised structure or subordinating verbal morphology.

Complementiser (dedicated particle introducing the clause) – e.g., English, German, Hindi, Esperanto, French, Danish, Maori, TokiPona, Arabic, Finnish, Russian, Spanish, Mandarin, Portuguese, Indonesian, VietnameseN, VietnameseS, Irish, Italian, Swahili, Bengali, Ewe, Georgian, Malagasy, Telugu, Tlingit, Tagalog, Thai, Persian, Hausa, Oneida
Nominalisation (the clause behaves as a noun phrase) – e.g., Japanese, Turkish, Hawaiian, Korean, Archi, Guarani, Nama, Pirahã, Navajo, !Xóõ
Subordinating verbal morphology (finite clause with subordinating suffix) – e.g., Basque, Wajarri, Greenlandic, Ubykh

Further reading:

Question 62: Quotative particles

Quotative particles mark quoted or reported speech and may also introduce complement clauses. This setting controls whether the language has a dedicated quotative marker.

Yes– e.g., Japanese, Hawaiian, Maori, Greenlandic, Russian, Turkish, Korean, Swahili, Archi, Bengali, Ewe, Georgian, Malagasy, Nama, Pirahã, Telugu, Tlingit, Thai, Hausa, Navajo, !Xóõ
No– e.g., English, French, Esperanto, Danish, TokiPona, Arabic, Wajarri, Finnish, Spanish, Hindi, German, Mandarin, Portuguese, Indonesian, VietnameseN, VietnameseS, Basque, Irish, Italian, Guarani, Tagalog, Persian, Oneida, Ubykh

Further reading:

Lexicon and writing

Lexicon and writing

These settings do not change the core grammar in the same way as phonology or syntax, but they strongly affect the visible flavour of the language. They control whether vocabulary is influenced by a prestige source and which script is used for displaying generated forms.

Loanword source

Loanwords can suggest cultural contact, prestige influence or a learned vocabulary tradition. The chosen source affects only borrowed vocabulary, not the genealogical identity of the generated language.

Question 63: Loanword source

This setting chooses the main cultural source for loanwords. It does not make the language genetically related to the source; it only shapes borrowed vocabulary and the cultural flavour of learned or prestige terms.

Latino-Greek (main source for European languages) – e.g., English, French, Esperanto, Danish, Russian, Spanish, German, Portuguese, Basque, Italian
Middle Chinese (main source for East Asian languages) – e.g., Japanese, Korean, VietnameseN, VietnameseS
Arabo-Persian (Islamic world) – e.g., Turkish, Arabic, Swahili, Persian, Hausa, Ubykh
Sanskrit (South Asian) – e.g., Hindi, Indonesian, Bengali, Telugu, Thai
English – e.g., Maori, Hawaiian, Wajarri, Finnish, Mandarin, Irish, Ewe, Nama, Tlingit, Tagalog, Navajo, Oneida
French – e.g., Malagasy
Russian – e.g., Archi, Georgian
None – e.g., TokiPona, Greenlandic, Guarani, Pirahã, !Xóõ

Further reading:

Writing system

The script setting controls how generated forms are displayed. It does not change the underlying phonology, though some scripts may approximate sounds that they do not traditionally represent.

Question 64: Script

This setting determines the script used for displaying generated forms. It does not constrain the underlying phonology, but some scripts may need approximations if the sound system contains features they do not traditionally represent.

Latin– e.g., Danish, Maori, Hawaiian, Irish, Japanese, Spanish, French, TokiPona, Greenlandic, English, Esperanto, Wajarri, Finnish, German, Turkish, Mandarin, Portuguese, Indonesian, VietnameseN, VietnameseS, Basque, Italian, Korean, Swahili, Bengali, Ewe, Georgian, Guarani, Malagasy, Nama, Pirahã, Telugu, Tlingit, Tagalog, Thai, Hausa, Navajo, Oneida, !Xóõ, Ubykh
Cyrillic [beta]– e.g., Russian, Archi
Devanagari [beta]– e.g., Hindi
Arabic [beta]– e.g., Arabic, Persian
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.