Skip to main content



Learning the phonology of a second language is a difficult task. Importantly, second languages (L2s) are often learned through both spoken and written input. L2 orthographic forms (spellings) can affect the perception and production of L2 speech sounds, and can lead to non-native-like pronunciation, especially in the early stages of L2 learning (Bassetti, 2008; Bassetti, Escudero, & Hayes-Harb, 2015). Perhaps the most obvious example is when L2 speakers pronounce silent letters, for instance producing comb (/kəʊm/ in British English) with an extra [b] (Bassetti & Atkinson, 2015). However, could orthographic forms even lead L2 speakers to perceive, produce, learn and identify speech sounds that do not exist in the L2?


The project aimed at investigating whether the number of letters in the spelling of an English sound affects the perception, production, learning and metalinguistic awareness of the sound in speakers of English as a Second Language.


The general question can be represented as follows:

<LL> → [Cː], [Vː]

That is to say, letter digraphs (<CC> or <VV>) will result in long sounds (geminate consonants [C:], or long vowels [V:]) in the perception, production, learning and metalinguistic awareness of EnglishL2 sounds of ItalianL1 speakers of EnglishL2. There were two predictions:

1) <CC> → [C:]. In English, consonant length is not contrastive. English consonants can be spelled with a single letter or with double letters, as in finish and Finnish. Both words are pronounced the same in English (/ˈfɪnɪʃ/). However, in languages like Italian, double consonant letters represent geminate consonants, that is to say long consonants, as in cane (/ˈkane/, 'dog') and canne (/ˈkanːe/, 'reeds'). If Italians recode EnglishL2 orthographic forms using the grapheme-phoneme correspondences of their native Italian language, they will produce and perceive a long [nː] in Finnish and a short [n] in finish.

2) <VV> → [V:]. In English, vowel length is not contrastive. English tense vowels are generally longer than lax vowels, although there are qualitative as well as quantitative differences. Long vowels are often spelled with double vowel letters, as in sheep /ʃiːp/, but can also be spelled with other vowel digraphs, including <V_e>, as in scene (/siːn/). Italian speakers often use length to distinguish English vowels, for instance producing ship as /ʃip/ (instead of /ʃɪp/) and sheep as /ʃiːp/. If Italians rely on English orthographic forms to determine the length of an English vowel, produce and perceive a long [iː] in seen and a short [i] in scene (both /siːn/).

Read more about Italian and English phonology and orthography and about the predictions here.

Approach and innovativeness

The project investigated this orthographic effect (number of letters → duration) from the points of view of perception, production, learning and metalinguistic awareness, as well as testing the effects of a pedagogic intervention. Our approach was innovative for the following reasons:

  • This was the first study to investigate orthographic effects on L2 phonology from all points of view -- perception, production, word learning, metalinguistic awareness, and teaching. Previous research focused just on production, or sometimes perception.
  • This is the first study of orthographic effects on L2 phonology to adopt an interdisciplinary approach. The team included an applied linguist (Bassetti), a psychologist (Masterson), a phonetician (Mairano) and a cognitive scientist (Cerni).
  • This was a large-scale study, which resulted in about 600 hours of recordings from 250 participants. The large number of participants afforded more complex analyses than in previous, typically much smaller-scale, studies of orthographic effects on L2 phonology.
  • The study investigated the perception and production of real words in experienced L2 speakers, including L2 learners with ten years' experience of L2 learning and bilinguals with long experience of naturalistic exposure. This contrasts with previous research which almost exclusively investigated orthographic effects on the early stages of L2 acquisition, looking at learners of novel languages, beginner L2 learners, or experienced L2 speakers learning pseudowords.approach


Data was collected from ItalianL1 learners of EnglishL2 in Italy, Italian-English late bilinguals in the UK, and English native speakers. Participants performed a variety of tasks tapping into their speech perception, speech production, metalinguistic awareness, novel word learning, as well as other relevant variables such as language proficiency. Over three hundred hours of testing were performed in Italy, with over a hundred learners of English in eight classes at three high-schools in Rome, where a teaching intervention was also run. In the UK, the project involved eighty Italian-English bilinguals who were late learners of EnglishL2 with long experience of living in the UK, and a control group of eighty English natives. This resulted in almost six hundred hours of recordings. Data analysis is ongoing.


Bassetti, B. (2008). Orthographic input and second language phonology. In P. Thorsten and M. Young-Scholten (eds.), Input matters in SLA (pp. 191-206). Clevedon, UK, Multilingual Matters.

Bassetti, B. & Atkinson, N. (2015) Effects of orthographic forms on pronunciation in experienced instructed second language learners. In R., Hayes-Harb, B. Bassetti, & P., Escudero (Eds.), Orthographic effects in second language phonology. Special issue. Applied Psycholinguistics, 36(1), 67-91. DOI:

Bassetti, B., Escudero, P. and Hayes-Harb, R. (2015) Second language phonology at the interface between acoustic and orthographic input. In R. Hayes-Harb, B. Bassetti, & P. Escudero (Eds.) (2015) Orthographic effects in second language phonology. Special issue. Applied Psycholinguistics, 36(1), 1-6.


Pronouncing an extra [l] in walk