In linguistics, the term native-level (near-native) speakers is used to describe speakers who have achieved "levels of proficiency that cannot be distinguished from native levels in everyday spoken communication and only become apparent through detailed linguistic analyses" (p. 484) in their second language or foreign languages. Analysis of native and native-level speakers indicates that they differ in their underlying grammar and intuition, meaning that they do not interpret grammatical contrasts the same way. However, this divergence typically does not impact a near-native speaker's regular usage of the language.
Domains of proficiency Although the vast majority of literature has shown that the age of acquisition of the learner is important in determining whether learners can attain nativelike proficiency, a small number of late learners have demonstrated accents and knowledge of certain areas of grammar that are as proficient as that of native speakers.
Phonetics and pronunciation Late learners who learn a language after the critical period can acquire an accent that is similar to that of native speakers, provided that they have attained relatively high levels of proficiency. In one study that employed speech samples of language learners, advanced learners of Dutch who spoke different first languages were tasked to read Dutch sentences. In addition, native Dutch speakers who were matched for levels of education served as controls. The recordings of these sentences from both the advanced learners of Dutch and the Dutch native speakers were played to Dutch native speakers (some of whom were Dutch language teachers). They were then asked to rate each speaker on a five-point scale based on how "foreign" they perceive the accent to be. The authors reported that four out of 30 learners were perceived by the native speakers to have achieved nativelike pronunciation. Similarly, the pronunciation of five out of 11 Dutch university students of English language and literature were perceived to be as good as native speakers of English even though they did not receive formal instruction in English before the age of 12. Second language learners’ pronunciation measured by voice onset time production has also been shown to be close to that of native speakers. Four out of 10 Spanish speakers who started learning Swedish as a second language at or after the age of 12 years exhibited voice onset time measurements that were similar to that of native speakers when asked to read out Swedish words with the voiceless stops /p/, /t/ and /k/. Likewise, a small number of late Spanish learners of Swedish were also able to perceive voiced and voiceless stops in Swedish as well as native speakers.
Morphology and syntax In the domain of morphology and syntax ('morphosyntax'), second language learners’ proficiency is typically tested using the Grammaticality Judgment Test. Participants in such studies have to decide whether the sentence presented is grammatical or ungrammatical. Similar to in the domain of phonology and phonetics, highly proficient second language learners have also shown near-native proficiency despite the fact that the target language was acquired at a later age. English, Dutch and Russian learners who exhibited near-native proficiencies in German were shown to perform as well as German native speakers in Grammaticality Judgment Tests that focused on word order and case markings. This was also consistent with an earlier study on learners of English as a second language who started learning English after puberty. These highly proficient learners displayed accuracy rates and reactions times similar to that of native speakers of English when asked to judge grammatical and ungrammatical wh- questions in the Grammaticality Judgment Test.
Common proficiency tests
Grammaticality Judgement Test The Grammaticality Judgment Test (GJT) is one of the many ways to measure language proficiency and knowledge of grammar cross-linguistically. It was first introduced to second language research by Jacqueline S. Johnson and Elissa L. Newport. Participants are tested on various grammatical structures in the second language. The test involves showing participants sentences that may or may not contain grammatical mistakes, and they have to decide on the absence or presence of grammatical issues. The test assumes that one's language proficiency is derived from language competence and language performance and reflects what sentence structures learners think are plausible or not in the language. However, in recent years, there have been papers that questioned the reliability of GJT. Some papers have argued that most sentences in GJT have been taken out of context. The lack of standardisation when administering the GJT in studies have also been deemed to be controversial, with the most prominent one being participants in untimed GJT performing better than those under timed GJT.
White noise test The white noise test was first developed by Spolsky, Sigurd, Sato, Walker & Arterburn in 1968. In these tests, recordings of speech with varying levels of white noise are played to participants and they are then asked to repeat what they have heard. Participants will need to rely more on their own knowledge of the language to understand the speech signal when there is a higher level of white noise in the speech signal. In addition, the ability of the second language speakers to decode what was being said in various conditions usually will determine their proficiency in the second language.
Cloze test Participants of cloze tests are typically given texts with blanks and are tasked to complete the blanks with appropriate words. To identify the correct item that fits the missing portion, cloze tests require second language speakers to understand the context, vocabulary, grammatical and pragmatic knowledge of the second language.
… excerpt ends here. Continue reading the full article.
