This measure is defined as:

a: The number of Aboriginal and Torres Strait Islander languages being spoken/used.

b: The number of Aboriginal and Torres Strait Islander languages categorised as strong.

Name: National Indigenous Languages Surveys (NILS) and the National Indigenous Language Report (NILR).

Frequency: Periodic. National Indigenous Languages Surveys are conducted at irregular intervals: NILS1 in 2004-05, NILS2 in 2014-15, NILS3 in 2018-19, and NILS4 in 2025-26.

Documentation (links): National Indigenous Languages Surveys | AIATSIS 

Provider name: Australian Institute of Aboriginal and Torres Strait Islander Studies (AIATSIS)

Provider area: AIATSIS Centre for Australian Languages

State and territory and Australia, by number of languages spoken/used currently.

State and territory and Australia, by number of new languages spoken/used currently.

State and territory and Australia, by number of languages categorised as strong currently.

State and territory and Australia, by number of languages that only have speakers/users aged 40 years or over and aged 60 years or over.

State and territory and Australia, by number of languages being revived/revitalised/reawakened.

Counting rules

For the NILS, AIATSIS collected data from respondents who self-identified as having sufficient knowledge of a language community in order to respond to questions about that community. These respondents include Elders, community leaders, Indigenous linguists, language centres, other organisations carrying out language projects, or non-Indigenous linguists who have close connections with the community.

Data for 2025-26 (collected via NILS4) uses the following definitions:

A language is defined as a distinct system of communication that is identified and described by communities. In all instances where languages were counted in NILS4, languages were categorised entirely according to guidance from communities. In previous NILS, this same methodology was used in most instances where languages were counted, but in some instances, a language was instead defined as a distinct system of communication that is not mutually intelligible with other languages/dialects. The Data quality considerations section advises which indicators may be comparable across surveys. 

A language is categorised as strong based on the five following criteria:

• The language has a high number of speakers/users (absolute number) - more than 1000 speakers/users, at any level.
• The language has a high number of speakers/users (proportion) - a majority of the language group speak/use the language, at any level.
• The language has multi-generational speakers/users - there are more than 250 speakers/users in each age group.
• The language is conversational - more than 500 people can speak/use the language at a conversational level.
• The language is used every day across all domains of life - on average, the language is spoken/used every day or almost every day in the following 7 domains: in conversation, when talking to children or babies, in ceremony, at home, in school, at work, in public.

These criteria reflect an understanding of language strength at a national level. Communities may understand the strength of their languages in different ways according to their local contexts and goals. In the case of languages that met most but not all of these criteria, AIATSIS contacted the respondent to discuss and determine whether or not the language should be categorised as strong, for example where a strong language is used by most people in a small community of less than 1000 people.

Data for NILS1 to NILS3 used the following definition: a strong language is used by all age groups, including all children, and people in all age groups are fluent speakers.

The following definitions are common across years (NILS1 to NILS4):

A new language is an Aboriginal or Torres Strait Islander language that has formed since 1788 from language contact between speakers/users of traditional languages with speakers/users of English and/or other languages. 

A traditional language refers to a language spoken by Aboriginal and Torres Strait Islander people prior to colonisation, or the directly descended varieties spoken today.

Languages being revived/revitalised/reawakened are languages which have had a break in intergenerational transmission but people are now learning, teaching or otherwise working to strengthen the language.  

Disaggregations:

• State and territory is classified according to the ‘geographical location’ of each language, as recorded on the AustLang dataset. (AustLang is a thesaurus of Aboriginal and Torres Strait Islander languages and peoples, maintained by AIATSIS.) The AustLang dataset records the state and/or territory of a language according to its cultural/historical location (using referenced sources). For further information see: AIATSIS AustLang Online . As the geographic location of some languages covers more than one state and/or territory, the sum of languages across states and territories, is greater than the national total.

Data from NILS4 may not be directly comparable with earlier NILS collections due to the differing samples of languages captured in each collection, and changes in methodology, including updates to how languages are defined and counted and how language strength is assessed. Differences may reflect methodological changes or sampling variability rather than changes in underlying outcomes.

Not all aspects of the aggregate data are comparable between NILS4 and prior collection. The number of languages categorised as strong is not directly comparable between NILS3 and NILS4. However, the number of languages spoken/used, the number of languages being revitalised, and by-language analyses may be comparable between NILS3 and NILS4, although sampling variability must be taken into account.  

Data for 2004-05 (collected via NILS1), 2014-15 (NILS2), and 2018-19 (NILS3) is provided for context in the Material for download. 

There are several factors needed to be considered when interpreting NILS results, both within a single NILS and over time:

• Each NILS has had slightly different objectives and therefore methodologies have differed. The methodology for future NILS may change, making it difficult to provide time series data.
• The term ‘languages’ can refer to distinct systems of communication that are not mutually intelligible. Some Aboriginal and Torres Strait Islander languages may be mutually intelligible but speakers/users describe them as different languages. For example, the Western Desert Language includes Ngaanyatjarra and Pitjantjatjara, which are mutually intelligible, but speakers/users tend to refer to them as separate languages. In NILS1 to NILS3, in the domain of language strength in particular, these languages may have been grouped together, whereas in NILS4 they were recognised as different languages in line with community understandings.
• Assessing and comparing proficiency of language users across the NILS series may be difficult because of the use of different measures of proficiency.
• It can be difficult to grade the vitality of a language.
• There are complexities in arriving at precise numbers. NILS respondents self report, and various respondents for an individual language variety may have differing interpretations of survey questions or may be considering differing criteria or measures in their response.
• NILS data may be affected by poor geographical and demographic coverage due to an inability to reach remote and rural areas, and an undercounting of children.

Other limitations of the data gathered include:

• AIATSIS received responses for 178 language varieties for NILS4. There are a number of languages which did not respond to the survey and as such have not data. The number of languages that were reported with speakers/users in 2025-26 was 177, while there was one language that reported zero speakers/users.
• In most cases, AIATSIS received one response per language variety. If there was more than one response, sometimes these answers conflicted. When this happened, the AIATSIS Centre for Australian Languages team made an informed decision about how to combine responses. These decisions considered the following principles: whether responses applied to part of or the whole language community, whether the community agreed on one representative response when contacted about it, whether respondents had different visibility of the language community/ies, and how respondents rated the certainty of their answers.

For more information on NILS3 see: AIATSIS - Third national Indigenous languages survey online: methods and limitations

Funding and timing for the next National Indigenous Languages Survey has not yet been determined.

Future reporting will seek to include the following additional disaggregations:

Languages by:

• geographic area (remoteness where possible)
• gender
• age.