THE ACOUSTIC PROPERTIES OF THE VOWELS OF PANGASINAN



THE ACOUSTIC PROPERTIES OF THE VOWELS OF PANGASINAN

Francisco C. Rosario, Jr.

University of the Philippines

217 Baldog, San Carlos City, Pangasinan, 2420

+639063550691

furashin@

ABSTRACT

Languages in the Philippines, though almost all belonging to the Austronesian language family, have unique ways in how sounds and sound patterns are realized. The number of phonemic sounds varies in each language.

The Pangasinan language, according to Richard Benton, has five significant vowel phonemes namely /a/, /E, /e/, /i/ and /o/ or /u/, and 13 consonant phonemes. The quality of these phonemes changes depending on the environment.

This study aims to use a scientific approach in describing the vowels of Pangasinan according to their acoustic properties and depending on the environments they are found. These environments include stressed and unstressed positions involving correlates like pitch, duration and intensity. A software program will be utilized to analyze the individual characteristics of these vowels.

In general, this study aims to contribute to the development of linguistic materials in teaching Pangasinan language to its native speakers, and even to interested second language learners.

General Terms

Languages

Keywords

Pangasinan language, vowel space, acoustic correlates, stress

1. INTRODUCTION

Each language has its unique characteristics. Syntactic structures, sounds and sound patterns vary in different languages even those belonging to the same language family. This explains why Philippine languages, even though showing similarities through cognates (lexical items showing similarities in form and meaning), have individual qualities.

These differences in every language entail something on the pedagogic aspect of language. For example, language learning and teaching may be greatly influenced by the differences in grammar and sound patterns. It is important, therefore, to investigate on the peculiarities of the languages, especially those which are included in the school curriculum.

2. OBJECTIVES

The researcher finds it very important to study the local languages in the Philippines. By doing such, knowledge on a particular language can be shared not only to the native speakers but also to other language speakers who are interested in learning the language.

In general, this research aims to provide a detailed description of the vowels of Pangasinan. For example, grammar books in Pangasinan have different claims on the number of significant vowel sounds, and it is the intention of this paper, by using a scientific approach, to clarify these problems that eventually lead to several issues in our understanding and appreciation of our language and even to language teaching.

The research also tried to give insights on the acoustic correlates of stress based on the acoustic properties (duration, pitch and intensity) of the vowels of Pangasinan which has been the central discussion of this paper.

3. SCOPE AND LIMITATIONS

In this paper, the researcher focused on Pangasinan language – a northern Philippine language being used and spoken by approximately 1.5 million of the population in the province of Pangasinan. This paper describes the vowels of Pangasinan by providing a vowel space chart and the acoustic properties of these sounds in stressed and unstressed positions.

Though this study provided insights on the acoustic correlates of stress, this does not give conclusive remarks on this problem. The results on this paper could be considered only as preliminary basis in identifying the factors contributing to stress.

The researcher had some limitations on the data used for this research. There were enough samples gathered for the vowels /a/, /E/, /e/, /i/ and /o/, However, the researcher only had limited samples for the vowel /u/.

4. REVIEW OF RELATED LITERATURE

Vowels, which are considered the nucleus of the syllable, are produced with less constriction in the vocal tract. These are also more sonorous compared to the consonants making them sound louder. This element of higher sonority has become the basis in forming the syllables. [McCombs 2006]

As stated earlier in this paper, different languages have different ways on how sounds are being articulated and produced. Two different languages can have different sound system and sound patterns.

According to Rev. Fidel of Amurrio [1970, 1-2], the Pangasinan language has four vowels /a/, /e/, /i/ and /o/ or /u/ and 14 native consonants. He stated that /o/ and /u/ are interchangeable. The use of these two sounds depends usually on the preference of the language speaker. For example, /u/ occurs in stressed syllables and /o/ for unstressed syllables.

Another scholar on Pangasinan grammar, Richard Benton [1971, 4], states that native speakers of Pangasinan recognize four significant vowel phonemes namely /a/, /e/, /i/ and /o/ and 13 consonant phonemes. He also added that the phoneme /E/ comes in for words of Spanish origin. For most Pangasinan speakers, /i/ is substituted to the vowel /E/ [10].

To further illustrate this, he provided a detailed chart of the articulatory properties of the vowels of Pangasinan.

Table 1. Vowel chart of English and Pangasinan vowels1

|LIPS |UNROUNDED |ROUNDED |

|Tongue |Front |Central |Back of |Back |

| | | |Central | |

|HIGH |/i/ [i1] |/i/ |[e] |/u/ [o1] |

|LOWER HIGH |/I/ [i2] | |[e] |/U/ [o2] |

|HIGH MID |[E] [i3] | | |[o3] |

|MID |/e/ [E] |/ǝ/ | |/o/ |

|LOWER MID | | |[a1] | |

|LOW |/æ/ |/a/ [a2] |[a3] |/ↄ/ |

In this vowel chart, we see the differences in how the vowels in English and Pangasinan are articulated. For Pangasinan, Benton indicated that a vowel phoneme like /a/ could have three variants or different ways of articulation – [a1], [a2] and [a3], so as with /i/ and /o/.

It is noticeable that Benton did not include /u/ in the vowel chart. He pointed out that /o/ is represented orthographically by the letters ‘u’ and ‘o’ which can be used interchangeably in words where /o/ occurs [14].

The occurrence of different variants of a phoneme can be attributed to the surrounding environment of a sound. Benton further noted that “many words in Pangasinan are differentiated only by the placement of the strong stress on the root… although the factors sometimes conspire to obscure such contrasts; the differences are observable and important in most environments in which the words concerned may be found” [28]. Because of these factors, non-native speakers of Pangasinan find it very hard to distinguish the position of stress.

Various studies have already shown that differences in the sound system of languages can greatly affect the teaching of a second language. The sounds present in the first language can be assimilated to the learning of a second language. Because of these differences in articulation, it is very important to provide a clear description of the distinctions of the sounds in a language for an efficient language teaching.

In a study conducted by Candalene J. McCombs (2006), she stressed on the idea that knowledge on the sound system of a particular language is very important in improving articulation and comprehension of a language. Her study included Colombian missionary students’ ability in learning English language. She mentioned that it is important for students to know how to discriminate the “acoustic boundaries” of sounds, however for learning to be more productive, language learners need to have more exposure and interaction to the target language users.

5. METHODS AND PROCEDURES

The participants in this study are six native speakers of Pangasinan (three males and three females) and ranged in the age from 19-39 years old (mean of age = 25.67).

The participants were asked to utter statements which carry in them the target words. The researcher first uttered the statements, and the respondents were asked to repeat the statements in their own way without considering how the researcher uttered the statements. Asking them to repeat the statements was preferred than asking them to read the same because the written form may even influence the pronunciation of words.

The statements were recorded using a software program called Praat. The target words were then extracted to measure the acoustic correlates, particularly pitch, duration (length) and intensity (loudness). The vowels were analyzed in both stressed and unstressed positions to see the acoustic properties of these sounds in these specific environments.

Formants (F1 and F2) were also measured and plotted using the JPlotFormants to see the vowel space of these vowels.

6. RESULTS AND DISCUSSION

Before discussing acoustic correlates of stress, it is first important to identify the vowel phonemes existing in a particular language. To do this, the formants (F1 and F2) of the vowels are measured and plotted to determine the “space” where these sounds are located.

6.1 The vowel space of Pangasinan

Based on the chart below, Pangasinan language has five vowel phonemes namely /a/, /E/, /e/, /i/ and /o/ or /u/. It is noticeable that the vowels /o/ and /u/ are located at almost the same location in the grid. From here, we can conclude that /o/ and /u/ in Pangasinan are not different at all. The articulation or production of these two sounds is identical.

Figure 1. The (stressed) vowel space of Pangasinan

[pic]

In stressed environments, /E/ and /i/ seems to be two different phonemes. However, there is not much difference when we consider these vowels in unstressed environments. The vowel space chart above indicates that /E/ and /i/ showed distinction. On the other hand, as shown below, /E/ and /i/ become identical sounds.

Figure 2. The (unstressed) vowel space of Pangasinan

[pic]

The changing behaviors of the vowels /E/ and /i/ and /o/ and /u/ are not surprising at all. As Benton already noted, speakers of Pangasinan language use these sounds interchangeably. Some may consider the mentioned sounds entirely differerent, however as we see here, these sounds are identical in position making them almost the same in production.

Using the figure above, we can conclude that the vowel /e/ is different from the English mid-central /ǝ/. The height of the Pangasinan vowel /e/ is higher, and is almost the same as the height of the vowels /E/ and /i/ and /o/ or /u/.

6.2 The acoustic properties of the vowels

6.2.1 Stressed vowels vs. unstressed vowels

One would normally assume that sounds in stressed position are longer compared to those in unstressed position. Using Praat TextGrid, we can visualize the distinction, in terms of duration, between the two sounds.

Figure 3. Praat TextGrid annotation of stressed vowel /a/ and unstressed vowel /a/ in the word anak ‘children’

[pic]

Aside from assuming the difference using perception as the basis or the visual representation illustrated above, it is necessary to prove scientifically if such distinction exists.

To do this, the researcher compared the duration values of the stressed vowels to the unstressed vowels. Based on the gathered data, it was proven that the duration of the stressed vowels and unstressed vowels show significant difference.

As shown in the data below, stressed vowel /a/ is significantly longer than the unstressed vowel /a/. This is also true to the vowels /E/, /e/, /i/ and /o/ or /u/.

Figure 4. Paired T-test of duration difference between stressed vowel /a/ and unstressed vowel /a/ 2

Paired T for D-strVa - D-unstrVa

N Mean StDev SE Mean

D-strVa 48 0.09572 0.04732 0.00683

D-unstrVa 48 0.06980 0.02026 0.00292

Difference 48 0.02592 0.04467 0.00645

95% CI for mean difference: (0.01295, 0.03889)

T-Test of mean difference = 0 (vs not = 0): T-Value = 4.02 P-Value = 0.000

Aside from duration, pitch and intensity were also considered to see if these acoustic elements contribute to the distinguishable properties of a stressed sound.

Pitch is not a significant element of stress in /a/, /E/, /ǝ/ and /i/. However, /o/ or /u/ considers pitch as a significant factor for stress, as shown in Figure

Figure 5. Paired T-test of pitch difference between

stressed vowel /o/ or /u/ and unstressed vowel /o/ or /u/

Paired T for P-strVo/u - P-unstrVo/u

N Mean StDev SE Mean

P-strVo/u 24 172.75 47.85 9.77

P-unstrVo/u 24 165.58 45.92 9.37

Difference 24 7.17 13.96 2.85

95% CI for mean difference: (1.27,

13.06)

T-Test of mean difference = 0 (vs not = 0): T-Value = 2.51 P-Value = 0.019

Intensity, too, is not as significant as duration in the vowels /a/, /E/ and /e/, but not for the vowels /i/ and /o/ or /u/. The intensity or loudness of these vowels is a distinguishing factor for stress.

Figure 6. Paired T-test of intensity difference between stressed vowel /i/ and unstressed vowel /i/

Paired T for I-strVi - I-unstrVi

N Mean StDev SE Mean

I-strVi 30 69.500 3.082 0.563

I-unstrVi 30 68.567 3.812 0.696

Difference 30 0.933 2.392 0.437

95% CI for mean difference: (0.040, 1.826)

T-Test of mean difference = 0 (vs not = 0): T-Value = 2.14 P-Value = 0.041

In this section, it is clearly shown that duration is the consistent and significant element that distinguishes vowels in stressed positions to the ones in unstressed positions. Pitch and intensity is not as significant as duration, however as shown in the data above, some vowels attribute stress to these factors.

6.2.2 Stressed vowels (penultimate) vs. stressed vowels (ultimate)

The comparison made earlier is not yet enough to make conclusions and generalizations regarding the significant elements contributing to stress. Sounds, whether in stressed or unstressed environments, can also be influenced by the position of the syllable.

In this section, we will take into consideration stressed vowels in penultimate and ultimate syllables.

It is important to take note that all vowels analyzed in this section have significant difference in terms of duration even though all are in stressed positions. The only difference is that vowels in the penultimate syllable are significantly longer than the vowels in the ultimate syllable.

Figure 7. Paired T-test of duration difference between stressed vowel (penultimate) /a/ and stressed vowel (ultimate) /a/

Paired T for D-strV1a - D-strV2a

N Mean StDev SE Mean

D-strV1a 24 0.13115 0.04076 0.00832

D-strV2a 24 0.06028 0.01712 0.00349

Difference 24 0.07087 0.04875 0.00995

95% CI for mean difference: (0.05029, 0.09146)

T-Test of mean difference = 0 (vs not = 0): T-Value = 7.12 P-Value = 0.000

Pitch is not significant in all vowels except /e/, which is indicated below. Pitch is significantly higher in the penultimate syllable.

Figure 8. Paired T-test of pitch difference between

stressed vowel (penultimate) /e/ and stressed vowel (ultimate) /e/

Paired T for P-strV1e - P-strV2e

N Mean StDev SE Mean

P-strV1e 12 182.2 52.1 15.0

P-strV2e 12 168.8 45.9 13.2

Difference 12 13.33 9.61 2.78

95% CI for mean difference: (7.23, 19.44)

T-Test of mean difference = 0 (vs not = 0): T-Value = 4.80 P-Value = 0.001

On the other hand, intensity shows significance only in /E/.

Figure 9. Paired T-test of intensity difference between stressed vowel (penultimate) /E/ and stressed vowel (ultimate) /E/

Paired T for I-strV1E - I-strV2E

N Mean StDev SE Mean

I-strV1E 12 72.33 4.33 1.25

I-strV2E 12 69.33 3.70 1.07

Difference 12 3.00 3.52 1.02

95% CI for mean difference: (0.77, 5.23)

T-Test of mean difference = 0 (vs not = 0): T-Value = 2.96 P-Value = 0.013

It was noted in the previous section that duration is the consistent and significant correlate of stress. In this section however, comparison between vowels in stressed positions indicates that stressed vowels in the penultimate syllable are significantly longer than the syllables in penultimate syllable.

6.2.3 Stressed vowels (V2) vs. unstressed vowels (V1)

All stressed vowels in the ultimate syllable, except /i/, do not show significant difference in duration. Pitch is significant in /e/ while intensity is not significant at all for the vowels included in this section.

Figure 10. Paired T-test of duration difference between stressed vowel (ultimate) /i/ and unstressed vowel (ultimate) /i/

Paired T for D-strV2i - D-unstrV1i

N Mean StDev SE Mean

D-strV2i 12 0.05061 0.01421 0.00410

D-unstrV1i 12 0.06606 0.02404 0.00694

Difference 12 -0.01545 0.02388 0.00689

95% CI for mean difference: (-0.03062, -0.00028)

T-Test of mean difference = 0 (vs not = 0): T-Value = -2.24 P-Value = 0.047

In this section, comparison between stressed vowels (ultimate) and unstressed vowels (penultimate) was made. The analysis shows no significant difference in duration.

The analysis made in this section blurs the idea that duration is the consistent and significant correlate of stress.

7. CONCLUSION AND RECOMMENDATIONS

Based on the data presented above, it can be concluded that Pangasinan language has five vowels - /a/, /E/, /e/, /i/ and /o/. It was clearly shown through the vowel space chart that /o/ and /u/ are in almost the same position. The chart also showed distinction between the vowels /E/ and /i/ in stressed position, however there is not much distinction in unstressed environment.

The Pangasinan vowel chart presented by Benton and is shown earlier in this paper, indicated /e/ in the high and lower high position. Benton’s vowel chart can be supported by the vowel space chart shown above. It is clear that the vowel /e/ is different from the English mid-central vowel /ǝ/.

This paper also discussed issues on stress by focusing on the acoustic properties of the vowels /a/, /E/, /e/, i, and /o/ or /u/ like duration, pitch and intensity.

In the first part of the analysis, duration shows to be the only consistent factor contributing to stress. However, comparison between two stressed vowels (one in the penultimate syllable and the other on the ultimate syllable), indicates that there is increased duration for vowels in the penultimate syllable. Also, the comparison between the stressed vowel (ultimate) and unstressed vowel (penultimate) disregards the idea that duration is the consistent factor.

Pitch and intensity can be significant factor but only for some vowels. It cannot be established based on the data gathered for this paper that pitch and intensity can become very significant factors contributing to stress.

Because of some limitations on the gathered data, this paper was not able to discuss fully the difference between the stressed and unstressed vowel /u/. Though it was already said that stressed /o/ and /u/ are located in almost the same position in the vowel space, it is still interesting to see the behavior of this vowel in unstressed position and compare it to /o/. By then, one can totally conclude if /o/ and /u/ really are the same or different phonemes.

This paper also tried to look at the problem on stress, however there was no intensive discussion and analysis made to explain the elements contributing to stress. The analysis was just limited to the acoustic properties of the vowels and on how these contribute to stress.

For now, duration seems to be the convenient answer for this problem, however, there are still reservations on whether to accept this or not. The problem on stress needs to be investigated further before any conclusion can be made.

The researcher suggests that intensive and comprehensive analysis on the acoustic correlates of stress be done on Pangasinan language. Other approaches can also be considered by looking at the syllables and not only limiting the analysis on the vowels.

8. NOTES

1The chart is taken from Richard Benton’s Pangasinan reference grammar [1971, 6]. The English vowel phonemes are in slashes and Pangasinan are in square brackets.

2To illustrate the results of the Pair T-Test, the researcher opted to include just one statistical result in the discussion if the other remaining vowel pairs show same results (statistically different).

9. REFERENCES

[1] Benton, R. 1971. Pangasinan reference grammar. University of Hawaii Press, Honolulu, HA.

[2] Ladefoged, P. 2003. Phonetic data analysis. Blackwell Publishing, USA.

[3] McCombs, C. J. 2006. The acoustic properties of vowels: a tool for improving articulation and comprehension of English. In  Forum on Public Policy: A Journal of the Oxford Round Table, Fall 2006, p.NA. Retrieved December 14, 2009, from Academic OneFile via Gale: .

[4] Rev. Fidel of Amurrio. 1970. Pangasinan grammar. Bugallon, Pangasinan.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download