Pronunciation model: Philippine English

View the key for Philippine English here.

The Philippines has immense linguistic diversity, with Ethnologue listing 183 living languages (Lewis, Simons & Fennig 2016), the largest native languages by number of speakers being Tagalog, Cebuano, and Ilocano (2000 census). Alongside these are the two official languages. English, introduced during the American annexation (1898-1935) largely replaced colonially-inherited Spanish and is the statutory national working language. Partnering English is the statutory national language of Filipino, based on Tagalog with vocabulary influenced by other regional languages. Proficiency in English has been deteriorating amongst all population groups, partially attributed to reduced use of English in school and increased used of Filipino in the media, yet globalization demands greater proficiency (Tayao 2008). The development of a distinct ‘Philippine English’ [PhE] alongside its General American English matrilect (Tayao’s term) has been documented since the 1960s. Although beyond the present scope to document its maturation (the reader is directed to Tayao’s description and the sources therein), a key feature of modern PhE is its primary variation appearing to be on sociological rather than geographical grounds despite the potential for different first-language interferences in different regions of the country. As with several of the varieties of English now reflected in the OED, PhE can be viewed as having three broad sociolects: a formal acrolect, a more heavily first-language-influenced (and hence more variable) basilect, and a widely-used and widely-accepted mesolect between them. The degree of observable American influence is greatest in the acrolect, with the mesolect and basilect demonstrating progressively greater differences.

The OED model for PhE is derived from Tayao’s (2008) descriptions of the various lectal forms, reflecting a mesolectal variety where possible but otherwise tending towards acrolectal rather than basilectal features.


The range of contrastive vowels is smaller than in the acrolectal form, but larger than in the basilect. Tayao describes mesolectal PhE as having no vowels distinguished by length alone (hence no length marks), and a number of vowels contrasted in acrolectal or American varieties are in free variation in the mesolect (at least for a significant portion of speakers, in spontaneous speech rather than careful and deliberate speech). Consequently, the OED model gives one vowel symbol to represent many groups for which most other varieties have a contrast, such as KIT and FLEECE. The STRUT vowel is initially given the symbol /ʌ/ by Tayao, but may be somewhat more of a stressed schwa as in the U.S. variety and Tayao also notes that it may be frequently substituted for /ɑ/ (the symbol chosen here as it is the most common vowel used in Tayao’s examples). The GOAT vowel /o/ is also in free variation with the two-quality (diphthongal) American /oʊ/. Most notably, the vowels of mesolectal unstressed syllables are consistently realized as ‘full’ vowels (as in the final column of the table above), unlike the acrolect. Tayao notes some word-specific differences also exist from American English, such as /o/ in the first syllable of model rather than the predicted /ɑ/, and off and of being homophonous.

Like its matrilect, PhE is a rhotic variety, with /r/ preserved after vowels in the NEAR, SQUARE, CURE, NORTH, FORCE, NURSE, START, and LETTER groups.

The stopping of /θ/ and /ð/ to /t/ and /d/ respectively is reflected in the OED. In actuality, it is another distinction in free variation in the acrolect and mesolect, primarily producing the fricatives in more careful and deliberate speech (unlike the basilect, in which they are consistently stopped).

Similarly, word-initial /z/ and /ʃ/ are reflected as such, but in word-medial position /ʃ/ becomes /sj/ while /z/ becomes /s/, and both become /s/ in final position (buzz, buses, rush). /ʒ/ does not appear distinct from /ʃ/, consequently following the same patterns (measure as /mɛsjur/). However, the affricates /tʃ/ and /dʒ/ exist as in American English. Syllabic consonants do not occur in the mesolect or basilect, and are rare in the acrolect, so are not shown in the OED pronunciations. Final consonant clusters of /s/ plus a consonant are reduced to the /s/ (as in fast), but more basilect-specific cluster reduction processes are not shown.

PhE does not exhibit the same stress patterning as U.S. English. Like HKE and SME, PhE also has a more syllable-timed rhythm than many other varieties, with each syllable having a similar duration. Levels of stress may therefore be perceived as less distinct than in American English. This is shown in the OED by a flattened stress structure, with all syllables transcribed as having at least secondary stress marks and with double-primaries permitted on multi-word entries. PhE also differs from U.S. English in specific word stresses. Mesolectal speakers appear to have a broad tendency to pronounce first-syllable-stressed words in U.S. English as second-syllable-stressed (e.g. colleague, baptism), and vice-versa (e.g. bamboo, percentage). Slightly more complex patterns (and even greater variability) appear to exist for longer words.

Several features were considered for inclusion in the OED model but are not reflected at the current time. The standard contrasts of /p/-/b/, /t/-/d/, and /k/-/ɡ/ in all positions, but it should be noted that word-initial aspiration of /p/, /t/, and /k/ in stressed syllables is not evident in the mesolect (this being a feature British and U.S. speakers would expect in words such as pear, tear and care). The glottal stops often used to ensure syllables start with a consonant is also not reflected, nor is the tendency of some mesolectal speakers to produce /b/ instead of /v/ (or, more rarely, /p/ instead of /f/). Different articulations of /r/ are also not distinguished.