LARRY E. SMITH AND CECIL L. NELSON
In recorded history, the present global spread and use of English is unparalleled (see discussions in Kachru 1986; Quirk & Widdowson 1985; Smith 1983; Strevens 1982). Crystal (1985) estimated that as many as two billion people have some ability in English. Alatis and Straehle (1997) cited a United States Information Agency (USIA) estimate of 700 million users of native and nonnative English, and also refer to English being “the most commonly used language at international conferences”; they also cite a British Council number of two billion users of English “with some awareness” of the language. Numbers cited and calculated by Kachru (2005: 14–15 and 205–207) indicate that English users in India and China alone number 533 million, a population of users “larger than the total [number of English speakers] of the USA, the UK and Canada.” Whoever’s figures are accepted, it is certain that the users of English in the Outer and Expanding Circles outnumber those in the Inner Circle.1 With such spread of the language, a frequently voiced concern is the possibility that speakers of different varieties of English will soon become unintelligible to one another. Bansal (1969) is an example of an early attempt to address this question. Citing Halliday, McIntosh, and Strevens (1964), Bansal wrote that “a very sensible view” was “that imported forms of English should be excluded [from consideration ‘for use as an educational model’] and mutual intelligibility should be attained by adopting ‘standard English grammar and lexis’ and keeping ‘the number of phonological units…close to those of other educated accents’” (Bansal 1969: 13). Van der Walt (2000: 173) wrote that “[t]he assumption that South Africans run the risk of becoming incomprehensible internationally was the motivation for [her study].”
In facing this question from a sociolinguistically realistic point of view, however, it must be kept in mind that for at least the last two hundred years there have been English‐speaking people in some parts of the world who have not been intelligible to other English‐speaking people in other parts of the world. Such is a natural phenomenon when any language becomes so widespread. It is not something that is “going to happen” but something that has happened already and will continue to occur. It is unnecessary for every user of English to be intelligible to every other user of English. Our speech and writing in English need to be intelligible to those with whom we wish to communicate in English. For example, there may be many people in India who use English frequently among themselves and who are not intelligible to English‐speaking Filipinos who also frequently use English among themselves; members of these two groups may not, as yet, have felt the need (or had the opportunity) to communicate with one another. These Indians and Filipinos may use English to communicate only with fellow countrymen and have little or no difficulty in doing so. If that is so, neither group needs to be concerned about its international intelligibility. Of course, there are many Indians and many Filipinos who use English to interact internationally, and they are the ones who must be concerned about mutual intelligibility.
Perhaps the concern about intelligibility can be rephrased in the following way: In international situations where people wish to communicate with one another in English, how intelligible are speakers of different national varieties? With the global spread of English, is the problem of understanding across cultures likely to increase in frequency?
Elsewhere (Smith & Nelson 1985), it has been argued that those who have traditionally been called “native speakers” are not the sole judges of what is intelligible, nor are they always more intelligible than “nonnative” speakers (Nelson 1992; Smith & Rafiqzad 1979). The greater the familiarity speakers, native or nonnative, have with a variety of English, the more likely it is that they will understand and be understood by members of that speech community. Understanding is not solely speaker or listener centered but is interactional between speaker and listener.
Understanding, or “intelligibility” in a broad sense, should be divided into three categories which make it accessible for examination and analysis in more specific terms:
Smith and Christopher (2001: 92–93) present an interactional scenario which serves to explicate these three components. An Australian woman is reported to have been having a conversation in English with a taxi driver in Istanbul. Things were going well “until she asked [the driver] to turn off the interior light”: the driver refused “sharply.” Since the passenger’s request seemed innocuous to her, and since a mutual compatibility in English had been established by the preceding conversation, she thought there had been a simple failure of intelligibility or comprehensibility – that the driver had misheard or misunderstood some part of her utterance – so she repeated it, only to receive a “near‐hostile” negative response and marked silence until the end of the trip, which terminated in the driver “almost [snatching] the fare from her and [driving] away rapidly.”
The failure of the interaction turned on a mismatch of interpretations of the female passenger’s utterance: she just wanted the light turned off, for whatever reason of comfort, or perhaps she simply thought it did not need to be on; the driver was not only culturally but legally bound not to allow himself to be in “a dark and confined space” with a woman. Smith and Christopher speculate that he may have been “shocked” by her request, which may have led to his responding as he did. Smith and Christopher (2001: 93) write:
The whole uncomfortable situation might have been avoided if the Australian had … [said], for example: “Do you always leave the interior light on when you drive?” Probably the driver would have replied: “I must leave it on, that is the law.” Another source of information may be a mediator, an independent third party familiar with [both cultures involved].
This example shows that successful communication in English is not assured by the participants exhibiting good pronunciation – the focus of so much attention in ESL and EFL teaching and learning – or even good lexis and grammar; utterances have pragmatic effects which cannot be interpreted without situational, social, and cultural awareness. These three categories – intelligibility, comprehensibility, and interpretability – may be thought of as degrees of understanding on a continuum of complexity of variables, from phonological to pragmatic, with intelligibility being lowest and interpretability being highest.
The remainder of this chapter reports on a study carried out by Smith (1992) that was designed to help determine (a) what differences, if any, there are in the intelligibility, comprehensibility, and interpretability of selected taped material of nine national varieties; (b) how familiarity of topic and familiarity of national variety influence the listener’s understanding of these varieties; and (c) whether the language proficiency of the speaker and/or listener influences the intelligibility, comprehensibility, and interpretability of these varieties. For this study, the nine national varieties, represented on tape, were spoken by educated speakers (at the graduate level at the University of Hawaii) from China, India, Indonesia, Japan, Papua New Guinea, the Philippines, Taiwan, the United Kingdom, and the United States. The tests of intelligibility, comprehensibility, and interpretability based on these recordings were administered to three different groups of native and nonnative educated users of English.
To test assumptions concerning the effects of proficiency in English and familiarity with topic and speech variety on understanding (intelligibility, comprehensibility, and interpretability), it was desirable to have both native and nonnative educated English users as subjects. These subjects exhibited a range in their degrees of proficiency in English and in their familiarity with the content of the selections, as well as in their familiarity with the national variety of English being used by the speakers. The three groups were composed of nonnative speakers, native speakers, and mixed nonnative and native speakers.
Group 1: Nonnative speakers. This group was made up of 10 nonnative English speakers from Japan whose English proficiency ranged from scores of 375 to 600 on the TOEFL test; four were students in the Hawaii English Language Program (HELP) at the University of Hawaii and six were students at the Japan‐America Institute of Management Science (JAIMS) in Honolulu. Subjects in this group were familiar with the Japanese variety of English, as well as with the content of the Japanese speaker’s presentation of “forms of address” (i.e. how Japanese address non‐Japanese in English at international meetings). Because they had studied English for at least 10 years and were students in the United States, they were also somewhat familiar with the American and British varieties of English and with the content of the US and British speakers’ presentations on “forms of address” (i.e. how British and Americans address outsiders in English at international meetings). However, these subjects were not familiar with any of the other speech varieties or with the topic of forms of address used in the other countries.
Group 2: Native speakers. This group was made up of 10 native speakers of American English who were undergraduate students at the University of Hawaii. All were quite familiar with the American English used by the American speaker on the tape, as well as with the content of her presentation. They were not totally familiar with any other of the speech varieties on the tapes but had had greater exposure to the Japanese and Filipino varieties than to any of the others. They knew little about forms of address in any country other than the United States.
Group 3: Mixed. This group was composed of one native and eight nonnative speakers, one each from Burma, China, Indonesia, Japan, Korea, the Philippines, Thailand, and the United States. Each of these people was fully fluent in English (having scored above 600 on the TOEFL test). As East‐West Center graduate students, they had all become familiar with several different national varieties of English. They were also familiar to some extent with the forms of address used in different countries because of their interactions at the East‐West Center with people from many parts of the world.
All three groups were balanced for age, sex, and educational background. The subjects with the lower TOEFL scores were highly intelligent and well educated, but they had not had much experience of interacting in English.
In order to have educated English speakers of the nine national varieties interacting with one another, graduate and postgraduate students at the University of Hawaii who were fluent in English were chosen to produce the listening passages. The speakers were asked to explain to an interactor who was of another national variety the forms of address used by people from the speaker’s country when they addressed outsiders in English. The respondent was, in each case, a person whom the speaker did not know and who knew little about, but was interested in, the speaker’s country. The respondent was instructed, in the speaker’s presence, to listen to the speaker, interrupt with questions of clarification when necessary, and give evidence of understanding the speaker by paraphrasing the important points the speaker made. The speaker was instructed to make sure the respondent understood how people in his/her country would address an outsider in English, for example, at international meetings both inside and outside his/her country.
Both speakers and respondents were told that the language in the recording session was to be informal but real. That is, they were to speak to each other as peers in an informal situation; they were not to pretend that they were other people or that the setting was another place. They were to recognize and accept the fact that they were two people in a recording studio at the East‐West Center in Honolulu. They were instructed that the conversation should stop once the speaker was satisfied that the respondent understood what the speaker had said about the topic. The interactors were told that they could make notes but that neither was to read directly from the notes. The sessions were unrehearsed, and lasted from 20‐40 minutes. The tapes of these sessions were edited down to 10 minutes of conversation which could be used as material for the comprehensibility and interpretability tests.
For the intelligibility test, the subjects heard a part of the conversation which was not used as part of the edited 10‐minute presentation. This was done so that the subjects would not hear any part of the conversation twice. In addition to the conversations with speakers of the nine national varieties mentioned above, one tape involving a speaker from Burma and a respondent from Thailand was also made to use in a demonstration of the testing procedure for all three subject groups.
Although no formal attempt was made to evaluate the difficulty level of the interactions, all were judged to be approximately equal in that (a) both speaker and respondent were fully proficient in English and believed themselves to be educated speakers of their national variety of English, (b) each person spoke clearly, and (c) the number of embedded sentences and the speed of delivery were approximately the same for all interactions. Of course, the setting and topic were always the same, and technical jargon was never used. In each case, if the speaker was male the respondent was female, and vice versa, so that on each tape both sexes were represented.
Three types of test questions were developed. A cloze procedure was used to test intelligibility (word/utterance recognition). Multiple‐choice questions were written to test comprehensibility (word/utterance meaning). Subjects were asked to paraphrase a small portion of the conversation they had heard in order to test their level of interpretability (meaning behind word/utterance). The test questions and directions were recorded by the same speaker.
Each testing session began with the researcher saying to each of the three subject groups that he was doing a study on the degrees of understanding of different national varieties of English and that he appreciated the subjects’ willingness to cooperate. They were assured that the results of the tests would have no effect on their academic work, but they were encouraged to do their best. The trial test was introduced, and the subjects were told that they could ask any question about the procedure during the sample test.
Each subject group then listened to the tape about forms of address in Burma and filled out the sample test items for the cloze procedure (intelligibility), multiple choice test (comprehensibility), and paraphrasing (interpretability). After the sample test, the subjects filled out the more subjective questionnaire (see Appendix), and they had the opportunity to ask questions about it. On that form, they were asked to state such things as (a) how easy/difficult it was for them to understand the speaker and respondent, (b) how much of the total conversation they had understood, (c) the nationalities of the speaker and the respondent, and (d) the English proficiency level of the speaker and the respondent.
Subjects then went on to the test proper. For each of the five paired recordings, each subject group first listened to the 10‐minute conversation, with the respondent asking questions and paraphrasing the important points. At the end of each conversation, the subjects were given a test that consisted of (a) a cloze procedure of a passage with 10 blanks (one at every seventh word) to be filled in as they listened, phrase by phrase, to a part of the original, longer conversation that they had not heard before; (b) three multiple‐choice questions based on the ten‐minute conversation that they had heard; and (c) three phrases taken from the 10‐minute interaction they were to paraphrase according to their interpretation of the meanings of the phrases in the conversation. This system was followed for each of the five paired recordings. That is, (a) the subjects heard a tape about a country, (b) the subjects were tested on that country, (c) the subjects heard the next tape about another country, and (d) the subjects were tested on that country. This continued until all five paired recordings had been heard and tested. The order of the five pairs of taped conversations was different for each subject group, to insure that any practice effect was balanced across varieties.
All of the tests to the three subject groups were administered on separate days within a two‐month period (October and November 1986). Identical playback equipment was used for each group, always in quiet surroundings. The tests were all graded by the same individual within a few days after they were completed and given to the researcher for analysis.
Tables 24.1, 24.2, and 24.3 present the tabulated results of the three parts of the test for each of the three groups. The speakers that the subjects heard are listed in alphabetical order (with respondents in parentheses) by country on the left side of each table. In each case, the percentage listed is the percentage of subjects in that group that answered 60% or more of the test items correctly. For example, from Table 24.1 we learn that, when listening to the speaker from the United Kingdom interacting with her respondent from Papua New Guinea, 70% of the nonnative subjects got 60% or more of the intelligibility test items correct, whereas 100% of the native‐speaker subjects and 100% of the mixed subjects got 60% or more of the intelligibility items correct.
The results tabulated in Tables 24.1, 24.2, and 24.3 show that all three subject groups did best on the test of intelligibility. All of the native subjects, all of the mixed subjects, and 92% of the nonnative subjects got 60% or more of the intelligibility test items correct. It appears that all of the interactions were highly intelligible to the three subject groups, but that the most intelligible were those with the speakers from Japan (respondent from China), India (respondent from the Philippines), and the United States (respondent from Indonesia). The pair with the speaker from China and respondent from Taiwan and the pair with the speaker from the United Kingdom and respondent from Papua New Guinea were rated as somewhat less intelligible across the groups. Language proficiency may have made a difference in the results of the intelligibility test, but being a native speaker was not shown to be a deciding factor, since the mixed group – with eight nonnatives and one native speaker – performed equally well on the test.
Table 24.2, concerning comprehensibility (word/utterance meaning), shows that the averages for all three groups were lower; 62% of the nonnative subjects, 82% of the native subjects, and 70% of the mixed subjects got 60% or more of the comprehensibility test items. The speaker from the United Kingdom and her respondent from Papua New Guinea were the most comprehensible, with 90% of the nonnative group getting 60% or more of the test items correct, and all of the native group and all of the mixed group doing the same. This is interesting, because this is the pair that was the least intelligible. This result shows that, as demonstrated in the Australian–Turkish example, the components of overall intelligibility or understanding are not necessarily dependent on one another in any specific case. It is also noteworthy that the Japanese speaker with the Chinese respondent was the second most comprehensible pair, with the pairs from India‐Philippines, US‐Indonesia, and China‐Taiwan being rated about equally in difficulty for comprehensibility.
Table 24.1 Intelligibility: subjects scoring 60% and above (NNS = nonnative speaker; NS = native speaker).
| Speaker (respondent) |
NNS: 10, all from Japan | NS: 10, all from US | Mixed: 9 (1 NS and 8 NNS, each from a different country) | Average % |
| China (Taiwan) | 90 | 100 | 100 | 97 |
| India (Philippines) | 100 | 100 | 100 | 100 |
| Japan (China) | 100 | 100 | 100 | 100 |
| United Kingdom (Papua New Guinea) | 70 | 100 | 100 | 90 |
| United States (Indonesia) | 100 | 100 | 100 | 100 |
| Average | 92 | 100 | 100 |
Table 24.2 Comprehensibility: subjects scoring 60% and above.
| Speaker (respondent) | NNS (10) | NS (10) | Mixed (9) | Average % |
| China (Taiwan) | 40 | 80 | 60 | 60 |
| India (Philippines) | 40 | 90 | 60 | 63 |
| Japan (China) | 80 | 80 | 70 | 77 |
| United Kingdom (Papua New Guinea) | 90 | 100 | 100 | 97 |
| United States (Indonesia) | 60 | 60 | 60 | 60 |
| Average | 62 | 82 | 70 |
Table 24.3 Interpretability: subjects scoring 60% and above.
| Speaker (respondent) | NNS (10) | NS (10) | Mixed (9) | Average % |
| China (Taiwan) | 40 | 60 | 89 | 63 |
| India (Philippines) | 10 | 40 | 78 | 43 |
| Japan (China) | 40 | 60 | 89 | 63 |
| United Kingdom (Papua New Guinea) | 10 | 50 | 89 | 50 |
| United States (Indonesia) | 30 | 40 | 100 | 57 |
| Average | 26 | 50 | 89 |
A priori, all of the subject groups might have been expected to comprehend the tapes about forms of address in the United States and the one about forms of address in the United Kingdom more easily than the others because all the nonnative members of each group had studied English for at least 10 years and had learned a great deal of cultural information about both countries. In other words, they knew the topic and were also somewhat familiar with each of these Inner Circle speech varieties. The Japanese group and the American group were expected to comprehend the tapes about forms of address in their respective countries more easily, since obviously they knew the information upon which the test for their country was based. A possible reason why the subjects failed to do this is that, although new information was given on each tape, the general topic was the same. This may explain why familiarity with topic was not a major factor in the subjects’ ability to comprehend the interactions. If the topics had been Noh theater, nuclear physics, or anything else besides forms of address, the effects of familiarity with the topic might have been greater.
Examination of the subjective questionnaires for each pair of interactors brought out other information which offered another possible explanation for the surprising lack of effect of topic familiarity. In both cases dealing with the American and Japanese subjects listening to a speaker from their country interacting with a respondent from another country, the responses of the subjects to the respondents may have been a factor. All of the native‐speaker subjects (i.e. the Americans) responded that they could easily understand the American speaker, but only 30% said that they could easily understand the respondent who was from Indonesia. Sixty percent of the nonnative group (i.e. the Japanese) responded that they could easily understand the Japanese speaker, but 70% said that they had some difficulty with the respondent who was from China. Their difficulty in understanding the respondent (e.g. “accent too heavy”) may have caused them comprehensibility problems with the overall conversation.
Table 24.3 provides some further insightful information. It is evident that the mixed subjects (one native and eight nonnatives, each from a different country) who had the greatest familiarity with different speech varieties were best able to interpret correctly the interactions of the five pairs of interactors. Twenty‐six percent of the nonnative speakers, 50% of the native speakers, and 89% of the mixed subjects were correct on 60% or more of the interpretability test items. The mixed subject group was better on all five pairs than were the native or non‐native subject groups. This is an important finding, which shows that interpretability is at the core of communication and is more important than mere intelligibility or even comprehensibility.
This part of the study offers supporting evidence that familiarity with several different English varieties makes it easier to interpret cross‐cultural communication in English. No doubt this facility is influenced by the fact that familiarity with different speech varieties also involves an awareness of cultural differences and some knowledge of various specific cultures. This is not to say that proficiency in the language itself is unimportant; the mixed group was fully fluent in English, although not at the native‐speaker level, except for one person. Additional evidence that proficiency is important is the fact that the native‐speaker subject group was better at interpreting all five interactions than was the less proficient nonnative subject group. The nonnative and native subject groups found the China‐Taiwan and the Japan‐China pairs easiest to interpret, and all three subject groups found the India‐Philippines pair the most difficult to interpret. Only the mixed group found the US‐Indonesia pair the easiest to interpret, perhaps because the mixed‐group members were the only ones familiar with many varieties of English. The native‐speaker group and the nonnative‐speaker group were not familiar with the Indonesian speech variety, and this may have been a factor in their inability to interpret the US‐Indonesia interaction correctly.
Other responses from the subjective questionnaires were also of interest. Table 24.4 shows the percentage of each subject group that thought they understood 60% or more of the conversations between the five sets of interactors. The mixed group of subjects, who were most familiar with different national varieties of English, had the most confidence in their ability to understand the conversations. All of the native‐speaker subject group thought they understood the US–Indonesia pair, and 90% of the nonnative subject group (i.e. the Japanese) thought that they understood the Japan–China pair. Familiarity with topic and familiarity with at least one of the speech varieties being used in a conversation apparently cause listeners to believe that they understand most of what they hear.
Table 24.5 shows the percentages of subjects making accurate guesses as to the nationalities of the pairs of speakers in the conversations. Again, the mixed subject group was best, followed by the native‐speaker group and the nonnative speakers.
Table 24.4 Percentage of subject groups that thought they understood 60% or more of the conversations between the five sets of interactors.
| Speaker (respondent) | NNS (10) | NS (10) | Mixed (9) |
| China (Taiwan) | 30 | 100 | 100 |
| India (Philippines) | 10 | 90 | 100 |
| Japan (China) | 90 | 90 | 100 |
| United Kingdom (Papua New Guinea) | 30 | 90 | 100 |
| United States (Indonesia) | 40 | 100 | 100 |
Table 24.5 Percentage of subjects making accurate guesses as to the nationalities of the pairs in the conversations.
| Speaker (respondent) | NNS (10) | NS (10) | Mixed (9) |
| China (Taiwan)* | 50 70 |
100 60 |
100 100 |
| India (Philippines)* | 20 60 |
90 100 |
100 100 |
| Japan (China)* | 90 80 |
100 100 |
100 100 |
| United Kingdom (Papua New Guinea)* | 70 10 |
70 20 |
89 100 |
| United States (Indonesia)* | 90 0 |
80 0 |
100 10 |
It was surprising, however, that even though 5 of the 10 interactors on the tapes identified themselves as nationals of a particular country, only the mixed group identified their nationalities accurately. (It was not surprising that few subjects recognized the Indonesian speaker, since they had been exposed to so few Indonesians. It would have almost certainly have been the same for the person from Papua New Guinea if he had not identified himself.) Another surprise was that the native‐speaker subject group was not as able to correctly identify their fellow American as the nonnative group or the mixed group was. The native‐speaker subjects were better able to identify the nationalities of the interactors from China, Japan, India, and the Philippines than they were the speaker from the United States. The nonnative subject group identified the speakers from Japan most easily and identified speakers from the United States more easily than they did speakers from the United Kingdom. Native speakers of English may be surprised to learn of the native‐speaker subject group’s low percentage of accuracy, and nonnative speakers may be equally surprised to see the nonnative subject group’s high degree of accuracy. The mixed subject group did well and, except for the Indonesian respondent, seemed confident in their responses. Speakers from all circles of English perhaps will be surprised that the mixed group did not guess the British person’s nationality with greater accuracy.
Responses to two other items on the questionnaire deserve comment. The questions dealt with the subjects’ perceptions of the interactors’ level of education and proficiency in English. The questions were identical except that number 7 was about the presenter and number 8 was about the respondent (see Appendix). The question was, “Based on what you heard, it seems that the presenter/respondent is (check as many as you wish): highly educated/educated/not well educated/a native English speaker/a nonnative English speaker/a speaker of Standard English/a speaker of non‐Standard English.”
Table 24.6 gives the tabulated results. First, a great majority of the subjects perceived the interactors to be in one of the top two categories, educated or highly educated. A small percentage (10%) of the nonnative‐speaker subjects thought the speaker from India was not well educated, and an equally small percentage (10%) of the native‐speaker subjects thought the respondents from Papua New Guinea and Indonesia were not well educated. Subjects in the mixed group thought all of the speakers and respondents were highly educated or educated.
Second, all three subject groups did well in correctly identifying the interactors as native or nonnative English speakers. The majority were accurate in every case except one: 60% of the native‐speaker subjects guessed that the respondent from Papua New Guinea was a native speaker of English. In spite of this, the native‐speaker group was able to label correctly more of the interactors as native or nonnative than either of the other two subject groups, with the mixed group a close second.
What is most interesting about Table 24.6 is the final listing, which concerns the subjects’ perceptions about whether the interactors were using Standard or non‐Standard English. Most of the native‐speaker subject group and the nonnative‐speaker subject group thought that everyone they heard used Standard English. The only exceptions were the speakers from India and Japan. Fifty percent of the nonnative subject group thought the speaker from India used non‐Standard English. The native‐speaker subject group was equally divided about the speakers from India and Japan in that 40% thought they used Standard English and 40% thought they used non‐Standard. It is heartening, from the world Englishes perspective, to learn that many native and nonnative speakers of English would label most educated speakers of nonnative English as users of Standard English. The mixed subject group was more critical and seemed to have a stricter criterion for Standard English: 33% of this group thought that the speaker from the United States used non‐Standard English. One might assume that many nonnative speakers and certainly most native speakers would label people who were clearly nonnative users as speakers of non‐Standard English. It seems clear, however, that nonnative English speakers need not be indistinguishable from native speakers in order to be judged as using Standard English.
Table 24.6 Subjects’ perceptions of interactors’ level of education and proficiency in English.
| Countries of interactors | HE/E/NEW % |
NS/NNS % |
SE/NSE % |
| NNS (10): | |||
| China Taiwan |
50/50/0 30/70/0 |
20/80 10/70* |
50/10* 50/20* |
| India Philippines |
30/60/10 50/50/0 |
0/100 30/70 |
30/50* 50/30* |
| Japan China |
40/60/0 40/60/0 |
20/70* 10/90 |
60/30* 80/0* |
| United KingdomPapua New Guinea | 40/60/0 50/40/0* |
70/20* 40/50* |
50/30* 40/20* |
| United StatesIndonesia | 50/50/0 10/90/0 |
80/20 40/60 |
70/20* 50/30* |
| NS (10): | |||
| China Taiwan |
70/30/0 40/60/0 |
0/100 30/70 |
70/10* 50/20* |
| India Philippines |
10/90/0 10/90/0 |
20/80 0/100 |
40/40* 50/30* |
| Japan China |
10/90/0 0/100/0 |
0/100 0/100 |
40/40* 70/10* |
| United Kingdom Papua New Guinea |
50/50/0 20/70/10 |
90/10 60/40 |
70/0* 60/20* |
| United States Indonesia |
40/40/0* 20/60/10* |
100/0 20/80 |
80/0* 60/30* |
| Mixed (9): | |||
| China Taiwan |
67/33/0 33/67/0 |
0/100 0/100 |
78/22 22/78 |
| India Philippines |
56/44/0 56/44/0 |
33/67 22/78 |
67/33 67/33 |
| Japan China |
11/89/0 22/78/0 |
0/100 0/100 |
11/89 50/50 |
| United Kingdom Papua New Guinea |
56/44/0 22/78/0 |
89/11 33/67 |
89/11 33/67 |
| United States Indonesia |
44/56/0 11/89/0 |
78/22 11/89 |
67/33 11/89 |
Key: HE = Highly educated; NNS = Nonnative speaker of English; E = Educated; SE = Standard English; NWE = Not well educated; NSE = Non‐Standard English; NS = Native speaker of English; * = Some subjects did not reply.
These results clearly support the interpretation that it is possible for Standard English to be spoken with many different accents. This is one of the very positive results of the vast spread of English across the globe.
In order to determine whether the spread of English is creating greater problems of understanding across cultures, this study was done with three types of subject groups (nonnative, native, and mixed) involving nine different national varieties. Understanding was divided into three elements: intelligibility, comprehensibility, and interpretability. Evidence supports the position that there are major differences between intelligibility, comprehensibility, and interpretability as defined in the study. Intelligibility (word/utterance recognition) is easier than comprehensibility (word/utterance meaning) or interpretability (meaning behind word/utterance). Being able to do well with one component does not ensure that one will do well with the others. Having familiarity with the information presented did not seem to affect any of the three groups, but those subjects having a greater familiarity with different varieties of English performed better on the tests of interpretability than did those who lacked such familiarity. Being familiar with topic and speech variety did affect the subjects’ self‐perceptions of how well they had understood. Language proficiency does influence intelligibility, comprehensibility, and interpretability, but it seems to be most important for comprehensibility. It is a striking result of this investigation that native speakers (from Britain and the United States) were not found to be the most easily understood, nor were they, as subjects, the best able to understand the different varieties of English. Thus, being a native speaker does not seem to be as important as being fluent in English and familiar with several different national varieties. These results indicate that the increasing number of varieties of English need not increase the problems of understanding across cultures, if users of English develop some familiarity with them.
Intelligibility has been a paradigmatic area of argument for those concerned about the cross‐variety acceptability of varieties of English since before the days of the term and concept of world Englishes. When Outer and Expanding Circle varieties (as we now think of them) are observed by some commentators in the Inner Circle, such as Quirk (1985), to cite a well‐known exemplar, concerns about imperfect learning are raised, and give rise to appeals to “interlanguage” and the righteousness of models. When those varieties are under scrutiny by some people, including policy makers, teachers, and linguists, in their own circles, similar apprehensions may arise: “Is our English serviceable enough for interaction in a world market?” (see Bansal 1969, and Nihalani, Tongue, & Hosali 1979). The work presented here shows that the rash response of attempting to teach and learn an Inner Circle variety in the Outer and Expanding Circles is, besides being a losing proposition, not a cogent answer, since not even all Inner Circle varieties of English are mutually intelligible with one another. More research of the kind reported on in this chapter and, for example, by van der Walt (2000) is needed, if these theoretical and pedagogical questions are to be directed usefully. Further investigations across varieties within the Inner Circle could help dispel the “us vs. them” mindset. And since all the evidence shows that most non‐Inner Circle uses of English across the world do not involve Inner Circle users, more studies of those interactions will continue to reveal what the criteria of intelligibility truly are.
Directions: Answer the following questions by putting a check mark (✓) in the appropriate space provided, according to how you feel about the taped material that you have just heard.
easily : with some difficulty : with great difficulty : not at all
easily : with some difficulty : with great difficulty : not at all
90%> : 75%–89% : 1%–74% : 50%–60% : 34%–49% : <33%
Yes _____ No _____ If Yes, check the appropriate reasons.(You may check as many as you wish.)
______ I could not understand the meaning of what was said.
______ One or both speakers spoke too quickly.
______ The accent of the presenter was too heavy.
______ The accent of the respondent was too heavy.
______ Other (please write) __________________________________________
highly educated _____ educated _____ not well educated _____
a native English speaker _____ a nonnative English speaker ____
a speaker of Standard English _____ a speaker of non‐Standard English _____
highly educated _____ educated _____ not well educated _____
a native English speaker _____ a nonnative English speaker ____
a speaker of Standard English _____ a speaker of non‐Standard English ____