Phonological features of Basilectal Philippine English: An exploratory study

This paper attempts to describe the result of a data-based investigation of the phonology of the Basilectal Philippine English as a response to Tupaz’ (2004) challenge to conduct Philippine English studies that would describe not only the “educated English” (the acrolect and mesolect speakers), but the “linguistic practices of genuinely marginalized voices (the basilect speakers) in Philippine society” (p.54), as described by Llamzon, 1997 in Tayao, 2004). The findings of this study provide a description of the phonological features of these “marginalized” voices that include minimally functionally literate Filipinos such as jeepney drivers, nannies, janitors, market vendors, and the like from a particular region and Visayan language variety – Cebuano speakers from Region 7 – to distinguish it from the previous studies that have usually sampled subjects of Luzon origin only. Finally, the paper echoes the call for future studies of Philippine English phonology describing the range of segmental and suprasegmental features of various Basilectal Philippine English speakers across the country.


Background of the study
As the additional and official language of Filipinos, Philippine English (PE) has its distinct characteristics, functions, and forms different from other World English like, for example, Singaporean English, Malaysian English and Thai English (Kachru, 1992).Moreover, its acceptance and legitimacy lie in the fact that English has penetrated the historical, functional, sociocultural, as well as the creative processes or contexts of the Filipinos (Kachru, 2004).Historically, language policies of the country have been formulated and revised time and again to accommodate the use of English in the educational system and to establish its place in such contexts.As an official language, English is used in various domains of function, which may include schools, mass media and World Wide Web, business and commerce, or government offices.The use of English in these domains contribute to the acculturation of English and its "native" speakers' ways of life, belief system, etc., into the Filipinos' psyche and culture-changing, transforming, or altering their sociocultural face or identity.In the same vein, the various literary genres, professional genres, and news media have been influenced by the conscious adaptation of the English language by the Filipinos.

Review of literature
As a legitimate and institutionalized variety of World English (Kachru, 2004), PE, particularly its sound system, has received considerable attention from scholars.Attempts at providing a description of the phonology of PE started with Llamzon's (1969) groundbreaking publiccation on PE, which he then entitled Standard Filipino English.After claiming that "there is a standard variety of English which has risen in the Philippines (and it) stands or falls short on the premise that there is a sizeable number of native and near-native speakers of English in the country" (p.84), he then sketched the structure of Standard Philippine English (SPE) based on the utterances of the representative speakers identified, alongside his identification of representative speakers of SPE and their norms of acceptability as well as his recommendation to target SPE in the teaching of English rather than General American English (GAE).His sketch of the structure of SPE primarily dealt with the phonology of the then purported (standardized) variety, hoping that someone would later on come up with a dictionary of Filipinisms, or "English expressions which are neither American nor British, which are acceptable in Filipino educated circles, and are similar to expression patterns in Tagalog" (p.46).
Amidst the criticisms that were thrown against the bold proposal of Llamzon (1969), it could be said that the study of and scholarship in the emerging variety of English in the Philippines came to be among the most pursued in linguistics in the Philippines.Needless to say, descriptions of the phonology of PE were also made by linguists  .Under the mentorship of Llamzon himself, Martinez (1972, 1975) wrote a (teaching) manual of SPE pronunciation.She did not have original data as basis for the manual; but rather, she used that of her mentor.A description of PE in the mass media was initially provided by Alberca in 1978 through his doctoral dissertation.In the same year, Gonzalez and Alberca came out with a publication about the same and Gonzalez, a little later in 1985, restated their descriptions in another publication.Their descriptions of PE in the mass media included listings of the distinctive phonological features.Casambre (1986), also made an attempt to characterize the features of PE in the mass media, including its phonological features, and likewise came out with the same distinctive features.His thesis, on the other hand, focused on distinctive phonological features per and across five periods -1901to1920, 1921 to1945, 1946 to1956, 1957 to1958, and 1968 to1983 -in the English language teaching history of the Philippines.Llamzon (1997) attempted to describe the phonology of the various groups of Filipino English speakers, which Regala-Flores 129 he categorized as Acrolect (considered formal and highstyle), Mesolect (falls somewhere between the prestige of the Acrolect and the informality of the Basilect; often the most widely spoken form of a language, generally being used by the middle class), and Basilect (typically differ from the standard language in pronunciation, vocabulary and grammar, and can often develop into different languages; a variety of a language used by people from a particular geographic area) following Strevens' (1982) and Platt and Weber's (1980) terms for the speakers' styles of talking (p.44).In his paper, he described the segmental features (production of vowels and consonant sounds) of the three groups of Filipino speakers vis-à-vis their American counterparts.The following are his findings: 1.The vowel and consonant sounds produced by Acrolect closely resemble those of the General American English phonemes and are clearly derived from them. 2. It is noticeable that the phoneme inventory of the Mesolect resembles that of the National Language (Filipino).Also, there is a tendency for the Mesolect speakers not to reduce unstressed vowels to the schwa.The STOP consonants were not aspirated in initial positions when followed by a vowel, and sometimes not released in final positions.
3. Among the Basilect speakers (Cebuano and Ilocano or Pangasinan, in his study), more substitutions are made than in the Mesolect for the Acrolectal phonemes.Llamzon (1997), attributed this to their ethnic tongue which forms the substratum.
Following Llamzon's (1997) group representative speakers of Philippine English, Tayao (2004), conducted a data-based study in an attempt to describe the distinctive phonological features shared in and between speakers of the three groups.The results of her study …suggest that a number of patterns of variation in the pronunciation of the three groups differ according to the social-group membership, thus supporting the notion of Acrolectal, Mesolectal and Basilectal norms for the PE accent and the study of PE phonology (p.86).
Her study also showed that among the Basilectal speakers, the vowel inventories among Cebuano and Visayan speakers showed only three vowels as utilized by the speakers coming from this group representative.She added, however, that among Basilectal Tagalog speakers, a five-vowel system would be realized and utilized.
In the recently published, The Handbook of World Englishes (Kachru et al., 2006), Bautista and Gonzalez (2006) Tupas (2004), however, posits that while these descriptive studies provide some insights into the phonological features of Philippine English, their overemphasis on Mesolectal and Acrolectal (or the so-called educated) speakers fails to give an adequate picture of the sound system of this variety of English.He laments this incomplete description in arguing that "by focusing simply on 'educated' English, studies on Philippine English have lent themselves towards elitist (socio) linguistics by almost completely ignoring the linguistic practices of genuinely marginalized voices in Philippine society" (p.54).These marginalized voices (that is, the Basilectal speakers in this study) include minimally functionally literate Filipinos such as jeep and tricycle drivers, nannies, janitors, and the like, whose speech patterns in English need to be described.This study was set to provide an initial description of the phonology of Basilectal PE, particularly the Cebuano speakers from Region 7 (Cebu, Bohol, Siquijor and Negros Oriental) residing in Metro Manila.The researcher hopes that the findings of this study may enrich the literature on the phonology of Philippine English as a legitimate variety of World Englishes.Answers to the following questions would be given, after an analysis of the English of some 48 Cebuano individuals: 1. How may Basilectal Philippine English be descry-bed in terms of its segmental features such as vowels and consonants? 2. How may Basilectal Philippine English be described in terms of its suprasegmental properties such as stress and intonation patterns?
This study investigated the speech patterns of Basilectal speakers of Philippine English.Speech samples of seven groups of exemplars of this variety of Philippine English will be taken: nannies/house helps, jeep/tricycle/pedicab drivers, hairdressers and salon staff, security guards, janitors, gardeners, street and market vendors, and department store sales ladies.Their speech samples were solicited using various techniques (see data collection section below for a detailed discussion).

METHODOLOGY The respondents
Given the exploratory nature and purpose of this study, which is to provide an initial description of the Basilectal Philippine English, this study only revolved around the description of the English as spoken by 48 subjects -20 male and 28 female.Most of them are ages between 23 to 32, but their age ranges from 18 to 52.All of them grew up in Cebu, Bohol, Negros Oriental and Siquijor and only moved to Manila later in their life (that is, after seven years old of age).Thus, all of them have Cebuano as their native language with Tagalog and English as additional languages acquired later in their lives.It should be highlighted here that the choice of Cebuano as the substrate language in the current study was deliberate: to distinguish it from the previous studies that have usually sampled subjects of Luzon origin only.This should allow for testing the possibility of language-specific influences if any, to the phonology of (Basilectal) PE.
The subjects use English only at work or in school and they claim average English proficiency across the four language macroskills.Details of their self-ratings of their English language proficiency are reported in Table 1.The subjects work as drivers, vendors, security guards, and household helps, among others, and the majority of them earn Php2, 100 -Php6, 000 per month.More than half of them completed the prescribed secondary education but a lot still did not, with only one being able to see but not finish college.These have qualified them to be Basilectal speakers of PE.After selection, the subjects were asked to read aloud a list which contains words and expressions that makes use of the critical segmental and supra-segmental features.A sampling of those words and expressions are found in the Appendix of this paper.The subjects were tape recorded while reading those words aloud.The resulting tape recordings were then transcribed following the IPA Phonetic Alphabet.

Data collection
The researcher replicated the data collection procedure used by Tayao (2004) and Llamzon (1997).The data were collected between January and April 2008 in the researcher's university and other major cities within Metro Manila (example, Manila, Parañaque, Quezon City, Pasig, Las Piñas and Mandaluyong).The data were gathered using a two-part instrument (see Appendix C for the Profile Sheet used in this study).The first part aimed to profile the respondents' personal information solicited their names, age, sex, province, age transferred to Manila, occupation, highest educational attainment, monthly income, and frequency of use of English in indicated domains (home, workplace, church, market, etc).It also revealed their own assessment of their English proficiency in terms of listening, speaking, reading and writing.People who did not meet the requirements are: 1) Cebuano-speaker from Cebu, Bohol, Siquijor and Negros Oriental; and 2) Transferred to Manila at least after their seventh year, were not considered as legitimate/qualified respondents.
The second part of the data-collection instrument elicited from the respondents examples of their spoken English, which were recorded on audio tapes.They were requested to read aloud a list which contains words and expressions that makes use of the critical segmental and supra-segmental features.Speech samples of the select seven groups of exemplars of this variety of Philippine English were elicited using the following techniques: 1) Oral reading of a list of words containing critical vowel and consonant sounds, 2) Oral reading of a list of words with "distinctive" stress placements, and 3) Oral reading of a structured dialogue for intonation pattern.
A sampling of those words and expressions are found in the  The researcher looked for trends across seven groups of respondents.Although an attempt was made to include a novel way to elicit spoken English from the respondents, an impromptu speech where respondents were to answer a speaking prompt provided by the researcher, this was dropped because the researcher observed that during the pilot testing of the datacollection instrument, respondents demonstrated discomfort, uneasiness, and embarrassment that resulted in prolonged silence and the expressed request/ decision not to participate in the research anymore.In view of this development, the researcher settled to adapt the procedures used by Tayao (2004) and Llamzon (1969) in their respective studies.

FINDINGS AND DISCUSSION
The findings of this study on the phonology of the Basilectal Philippine English are presented in two sections, the first is the description of their segmental sounds, and the second is the description of their suprasegmentals, with particular reference to word stress and intonation patterns of Basilectal PE.

The segmental analysis of Basilectal PE
The following are summary of the findings of this study at the segmental level: The consonants Language Samples Project ( 2001)).The aspirations of the GAE voiceless stops (/p/, /t/, /k/) especially in the initial and final positions was absent among the respondents, example, the words Paul, toe and look were realized without aspirating the sounds of /P h /, /t h /, and /k h / in the initial and final positions, respectively.Likewise, the GAE voiced stops (/b/, /m/) in bamboo and map are not present in all occurrences.
1.The voiceless /f/ and the voiced /v/ in five and vibes also known as labiodental fricatives, occurred in a clear majority of cases except in one instance (that is, the word vote) where half of the respondents replaced /v/ with the bilabial stop /b/. 2. The GAE interdental fricatives /θ/ (voiceless) and /ð/ (voiced) was absent among the respondents and are replaced by the alveolar stops /t/ and /d/ respectively such that the /θ/ phoneme in thank is pronounced as /t/ and the /δ/ phoneme in those is pronounced as /d/.
3. The GAE alveolar fricatives /s/ and /z/ are realized as /s/ in all instances regardless of their positions: initial, middle, and final, example, as in zoo, thousand and buzz.4. Majority of the respondents realized the GAE affricates /š/ and /č/ both in the initial and final positions occurred in a clear majority of cases (example, ship, sheep, shepherd, church, chart, watch and touch).5. Compared to the GAE retroflex /ɽ/, respondents rendered it as a rolled or one-tap /r/, example as in fortyfour.
6. Majority of the speakers realized the consonant clusters in the initial position, (for example, twelve, shepherd) and in the final position, however, only when the words end in /rk/, /nk/, /rd/, /nd/, /lv/, as in words like, fork, tank, shepherd, thousand, twelve; but never in consonant clusters that end with the sound of /t/ (e.g., kissed, perfect) which were realized as /kɪs/ instead of /kɪst/ and /pɛrfɛk/ instead of /pɛrfɛkt/, respectively.
Note the deviation from some GAE consonant inventories particularly in the production of fricatives at the labiodental, interdental, and alveolar points of articulation.
It may be worth mentioning here that the Basilect Cebuano respondents in this study do not differ from their Mesolect and Basilect counterparts in Llamzon's study (1997, p. 46) and Tayao's (2004, p. 82) in terms of consonant inventories.This could be attributed to the fact that Filipinos in general do not tend to aspirate these STOP consonants (that is, /p/, /t/, /k/), substitute /t/ for /θ/ (voiceless) and /d/ for the (/ð/ (voiced), among other substitutions.Llamzon (1997), stated that although "Filipinos are willing to copy GAE, they retain something of their identity-in their lack of nasal twang, in the careful articulation of individual syllables, and in their refusal to use the 'reduced signals' of the informal conversational style of GAE" (p.43).
In the same vein, one can argue that Filipino speakers, be they Acrolect or Basilect, at some point, decide not to follow or speak like a 'native' American so long as they can be understood or are able to communicate their ideas, feelings, or desires.This is also true for other Asian speakers of English.The phonological system of the General American English or even the British English serves as a guide and is not meant to be strictly mimicked or aimed at.However, a closer look at the consonant inventories of the respondents in this study Figure 3. General American English (Adapted from those of American English used by Ladefoged (1995) and The Language Samples Project ( 2001) confirms a general notion that lack of training/exposure to the language may have greatly contributed to their inabilities to produce the sounds under study.It may be worth reiterating here that most of the respondents are high school graduates.Nevertheless, given the number of years they have been exposed to the English language (ten years, if we go by the Department of Education's Bilingual Policy in effect now), it is still sad (to say the least) that these phonological features have not been mastered or at least learned by them.Of course one can always argue that their ethnic tongue forms the substratum which is responsible for the substitutions or mis-production of these sounds.

The vowels
The chart presented in Figure 3 and 4 below are intended to represent the vowel system of the Cebuano speakers studied in this study: 1.There is a series of vowels at the FRONT upper high /i/, lower high /ɪ/ higher mid /e/, and mid /ɛ/ tongue heights, e.g., as in please, dip, gate, and lend, respectively.2. There are two CENTRAL vowels at mid /ʌ/ and low /a/ tongue positions realized in such words as truck, and track.3.There is a series of vowels at the BACK upper high /u/ (example, put, look, full), lower high /ʊ/ (example, to and Luke), higher mid /o/ (example, vote, bought, and mango), lower mid /ɔ/ tongue positions such as Paul, saw, and call).It is interesting to note that this group of respondents for Basilect PE speakers, yielded different results.Compared to previous studies conducted where the representation of the Basilectal vowel system of many Cebuano and Visayan speakers of English are described as utilizing "only three vowels" (Cf.Tayao, 2004, p. 84;Llamzon, 1997, p. 47), there is a noted production or realizations of other vowel sounds as in the case of /ɛ/ and /ɔ/ in words like lend and Paul; /e/ and /ʊ/ in instances like gate and to .Although not fully or distinctly realized, the results of this study indicate that there are occurrences where speakers were able to produce the ten vowel sounds.
Realizations of these variants of vowel sounds could have been made possible by the exposure to American music, movies, and other forms of media to which all respondents confirmed listening to or watching.Likewise, news reports over local channels could have been responsible for these productions, for although majority of the news reports are now in Filipino, presence of English words, phrases, and idiomatic expressions are interspersed in the news.The advent of text messaging could also be an intervening factor for respondents admit passing on or forwarding English quotations they regularly receive.

The suprasegmental analysis of Basilectal PE
The suprasegmental features of phonology given emphases in this study are stress and intonation.The following are the detailed explanations for each.

Stress
The investigation of the word stress among the Basilectal PE speakers for this study was made possible through the oral reading of a list of words with "distinctive" stress placements compared to GAE.The second part of the data-collection instrument facilitated elicitation of these data via audio recordings which in turn were transcribed and analyzed.Table 2 shows the list of these words, the syllabication per word, and the placements of the primary stress.Based on the data shown in Table 2, the following summaries are arrived at: 1. Of the two two-syllable words with GAE stress on the first syllable, only one, carton, received the initial word stress from the majority of the speakers.The primary stress was placed on the second syllable of the word, menu.
2. However, the lone two-syllable word, bamboo, with the GAE stress on the second syllable was a variation where the Basilect speakers placed the stress on the first syllable.
3. Of the three three-syllable words with GAE primary stress on the first syllable (example, talented, seventy, broccoli), the group diverged from the GAE pattern in the case of talented and broccoli.However, majority of the speakers placed the primary stress on the first syllable of the word seventy in accord with GAE pattern.4. In the case of three three-syllable words where GAE stress is placed on the second syllable, majority of the speakers conformed to the GAE pattern of placing the stress on the second syllable of the words determine and utensils.However, with the word, percentage, the group diverged from the GAE pattern by placing the stress on the initial syllable.5.In the case of seven four-syllable words where the initial syllable receives the primary stress in the GAE pattern, majority of the words received primary stress on the second syllable as in the case of honorable, cemetery, comfortable, category and ceremony.In the case of the words military and elementary, the speakers placed the primary stress in the third syllable of the said words.Interestingly, the word elementary was read by all speakers as having five syllables compared to the four-syllable word in the GAE syllabication pattern.
The preceding results seem to indicate that generally, the Basilectal PE stress pattern differs greatly from its GAE counterpart, as seen especially in the cases where three-and four-syllable words like percentage, honorable, cemetery, ceremony and elementary.However, this study revealed a slight change in the placement of primary stress on the second syllable of the word utensil where the Basilectal group in other studies placed it on the first (cf.Tayao, 2004, p. 85).This only suggests that until further comprehensive studies are conducted with regard to stress patterns of Basilectal speakers, no conclusive claims can be made in relation to this suprasegmental feature of PE phonology.

Intonation
Intonation is the other suprasegmental feature investigated in this study.In order to describe and analyze the intonation pattern of the seven exemplars of the Basilect speakers in this study, the second part of the datacollection instrument elicited sample speech patterns of the respondents via oral reading of a structured dialogue (see Appendix E for a copy of this).These were audiotaped, which in turn were transcribed and analyzed.The findings of this study may be summarized as follows: 1. Generally, majority of the speakers in this study follow the GAE rising-falling intonation pattern where the voice of the speaker begins with the normal tune (2) and ends by raising it to high (3) on the last stressed syllable or word of the sentence before making it fall to low (1) at the end.In the case where the speaker in the structured dialogue had to state a fact, the respondents followed the rising-falling intonation pattern required in expressing simple statements of fact as in the following examples: Oh, yes, Miss.
I think I have met you before.
2. However, the respondents diverged from GAE intonation pattern where request for information, which required 2-3-1 intonation pattern, occurred.All the respondents remained in the high tune at the end of the question, as in the cases below: What's your name?
How about you?
3. In the case of yes-no question, findings reveal that majority of the respondents raised their voice at the end of the question in conformity with the GAE intonation pattern as in the following instances: Are you from Cebu? Do I?
We used to be classmates, remember?
Based on the preceding discussion, it suggests that on the discrete-point or micro level, the intonation pattern of the Basilectal group investigated in this study does not vary from the GAE intonation pattern, especially in the realizations of the general rules discussed above (example, rising-falling intonation where simple statement of Regala-Flores 135 facts, commands or requests, or request for an information involving the wh-questions and the rising-rising pattern involved in yes-no questions are concerned).However, analyzed from a global or macro level perspective, the transcribed data would reveal that the respondents in this study do not fully reflect nor realize the GAE intonation pattern because of the staccato or disjointed reading of the structured dialogue.The natural flow of tones and the constancy of the rise and fall of the voice among the Basilectal group may not fully reflect the GAE intonation pattern as a whole.What makes the findings of this study interesting is the fact that they were able to raise and lower their voices in the instances cited above.Again, this only suggests that further studies be done before a set of intonation patterns characteristic of the Basilectal PE phonology be established.

CONCLUSION
This paper has attempted to provide an initial description of the phonology of the Basilectal Philippine English focused on the Cebuano speakers living within Manila and its surrounding cities.In response to the challenge posed by Tupas (2004) with regard to the incomplete description, if not a dearth of studies, involving "the linguistic practices of genuinely marginalized voices in Philippine society" (p.54), such as jeepney and tricycle drivers, nannies, janitors, and the like.However, due to certain limitations and constraints, the researcher concentrated on Basilectal speakers from Region 7, consists of Cebu, Bohol, Negros Oriental and Siquijor only.The decision was based on the perceived notion that Cebuano and Visayan speakers pose a very distinct phonological characteristics compared to Tagalog speakers.Aside from the specified place of origin or region, the study further limits its selection of respondents to factors like age, occupation and level of proficiency.This is to ensure that strict adherence to the Basilectal characteristics delineated by Llamzon (1997, in Tayao, 2004) is followed.
In the presentation of the results of this fieldwork investigation, a simple quantification of the analyzed data was used: simple calculations of the frequencies of occurrences.Findings from this study seem to suggest that at the segmental and suprasegmental level, the Cebuano speakers from Region 7 do not diverge much from the GAE pattern.These results seem to reveal (if not challenge) certain phenomena not observed in previous findings of other studies (Cf Bautista and Gonzalez, 2006;Tayao, 2004) such as non-realizations of the interdental fricatives /f/ and /v/ and divergence from GAE stress placement of some words among their respondents.These developments in the "evolving phonology" (to borrow Tayao's term) of the Basilectal PE make this field more exciting and inviting for future research not so much towards the standardization of the Philippine phonology but towards a more empirical-based analyses and description of the various phonological features of the various geographical and linguistic backgrounds of PE speakers.May this paper be a contribution to more detailed analyses/studies of the phonology of Basilectal Philippine English.

Figure 1 .
Figure 1.General American English

Figure 1 Figure 2 .
Figure1and 2, schematically represent the results of this study in relation to the production of consonant sounds realized by the respondents.(The consonant and vowel charts used in this study are adapted from those of American English used byLadefoged (1995) and The

Table 1 .
Self-ratings of the English proficiency of the respondents.

Table 2 .
List of words, syllabication per word and the placement of primary stress.