Lexical Variation in the Rural North Jordanian Dialect

This study investigates lexical variation, which is due to more education, more mobility, and widespread use of the social media, in the dialect of three towns around Irbid City in north Jordan and its correlation with age, gender, and level of education. Labov’s approach is adopted to examine the linguistic variation among 98 speakers of the Irbidite dialect. Around 100 words were collected and put in the form of a questionnaire to elicit the opinion of speakers from different age groups, genders, and levels of education towards the frequency of their use of these words. The study used the method of direct interview to elicit the feelings of the participants about the dialect they use. The results show that old speakers and less educated ones tend to preserve their native lexical items more than others. They indicated that they use the original lexical items because they are proud of their dialect which reflects their identity. The groups which tend more to neglect some lexical items are educated young and middle-aged female subjects. They indicated that they do so for prestige and imitation of peers in the Irbidite society.

marker is a variable that connects to stylistic variation as well as class, sex and/or age . Speakers are subconsciously aware of it ,as they use a variant or more in formal style and one or more variant in informal speech style. In Labov's New York study, the /r/ variable has a social stratification marker. In the event that a speaker articulates the [r], at that point he is from the high class, yet in the event that not, at that point he is from the lower class, as in these models, [ha: rd]  The second variable, the indicator, is considered to be socially marked when it varies with social attributes. For example, some speakers in North America make a distinction between vowels in "Ben" pronounced as "bun", "buses" pronounced as "bosses", while others do not. Labov gives the merger of vowels in "hawk" and "hock", as an example of indicators. The level of the merger of vowels is not the same across people, yet is regularly underneath the conscious awarenes. Finally, a stereotype is the most marked variable and can be changed in the course of time. Consequently, the linguistic variable moves from the category of marker to that of the indicator and vice versa. A stereotype often becomes avoided because it is stigmatized.
Most scholars (e.g., Trudgill, 1972;Gray, 1998;Al-Wer, 2000;Fought, 2004) believe that social factors like age, gender, and level of education have an impact not only on language use in general but also on lexical choice, as speakers may change elements of their language several times during their life, and females generally speak differently from males, and so do well-educated and less educated people.
At the same time, social media like Twitter and Facebook seem to have triggered lexical variation and innovation in many languages such as English. Robinson (2019), for example, studies lexical variation across the UK focusing on region and jargon, and Grieve et al. (2019) investigates lexical dialect variation in British English using Twitter. Grieve (2016)

handles regional variation in written American
English. Moreover, Grieve et al. (2017) and Grieve et al. (2018) deal, respectively, with the emergence of new vocabulary in Modern American online ad with lexical innovation in American social media.
For his part, Austen (2017) compares Black with White regional variation in the USA. Different variations among the speakers of the rural dialect of three towns at the eastern edge of Irbid City appear at levels associated with variables such as age, gender and education. Besides phonetic variations, one especially notices lexical differences between the old and young, men and women and well-educated and less educated. These lexical differences are investigated in relation to age, gender, and education. The three towns which represent the locale of the study are Huwwarah (36,000 inhabitants), Bushra (28,000) and Sal (16,000) and they occupy an area of about 36 km2 (Statistics obtained from Greater Irbid Municiplaity, 2020). The original inhabitants of the three towns all speak the type of Jordanian Arabic known as the rural sub-dialect, usually differentiated from the so-called Urban and Bedouin sub-dialects (see Al-Khatib, 1988

Theoretical Framework and Literature Review
The study draws on Labov's approach since it focuses on studying the relationship between language and the society which uses it. Since the 1960s, Labov's work has introduced a groundbreaking approach to investigating the relationship between language and culture, creating an area that has been called "variationist sociolinguistics" and maintaining the linguistic structure inherent in a core theory of this field. How a language is spoken (and written) varies between people and situations faced by the same person. Labov argued that these variations are essential to the functioning of a language and are not only common but also necessary. This interpretation contradicts much of linguistic theory's historically prevalent thought and practice, from Ferdinand de Saussure to Noam Chomsky.
Mainstream theoreticians do not deny that there is a variation but tend to diminish its relevance and consider it as a superficial phenomenon which obscures a fundamental uniformity that characterizes language. Labov's research shows that linguistic variations can be generalized and highly structured and that they reveal regular co-occurrence patterns between language forms such as the pronunciation of a certain vowel and socioeconomic classes. Such insights derive from a socially realistic perspective, which takes into account how a variety of speakers in day to day situations use the language. Labov advocated a stronger empirical basis for linguistics, called into question the relevance of analysis based on a native speaker's insights, and emphasized the value of observing speech that has been naturally generated. His approach is characterized by his focus on quantitative methodologies and others within sociolinguistics. Often only with the statistical analysis, patterns of co-variation between linguistic forms and social variables are apparent. During his career, Labov explored a range of language phenomena in the paradigm of variation. He also studied semantical (e.g., quantifiers each and all) and grammatical characteristics (e.g., copulation contraction and deletion) but predominated in his work on phonological variation. Labov has given significant attention to issues of the language change in addition to investigating the synchronic dynamics of social-linguistic variability. Most of his research studies English and he has played an important role in US dialectology, where he helped to deprive scholars of their traditional focus on maintaining regional patterns of expression.
Regarding variation in English, many studies have been conducted. For example, a study conducted by Warkentyne (1972)  the speech of old males and females, followed by the middle-aged males and females, and then the young males. It was found that young females abandoned the local dialect and used the urban one. Palva (2008) surveyed two spoken dialects in Jordan: Salti (west of Amman) and Karaki (100 kms south of Amman).
He found that, despite many similarities in the two varieties regarding phonetics, morphology and syntax, the Salti dialect exhibited many features of "Syro-Mesopotamian" rural Arabic, while the Karaki dialect had shared characteristics with the rural and bedouin dialects of the Sinai and the Negev as well as the Jordanian Bedouin dialect to the East of the Gulf of Aqaba.
El-Salman and Roche (2009) investigated the speech of Tirawi people who emigrated from Haifa (Palestine) to Irbid due to the Arab-Israeli war in 1948. The study dealt with the influence of migration on the lexical items in Tirawi dialect. It discussed how migration led to the abandonment of traditional farming practices, which in turn led to lexical items associated with that lifestyle disappearing from use.
The results show that old people prefer to preserve their native lexical items when they talk about farming. On the other hand, the middle-aged and young groups use the Tirawi lexical items at a very low rate. Prestige did not play a central role in the variation. In sum, the study showed that when a person migrates from one place to another and leaves his traditional practices, the lexical items relating to those practices disappear.

Methods and Procedures
The sample of the study consisted of 98 (44 males and 54 females) native speakers of Irbidite dialect from the three towns under investigation. In detail, it consisted of three main age groups: the old group (55 and above), the middle-age group (35-54) and the young group (18-34) with various levels of education. The distribution of the participants is shown in Table 1   The data were collected using three complementary methods. The first method was observation.
Through contact with relatives and friends who live in Huwwarah, Bushra and Sal, it was easy to gain a large amount of data and take notes from the interaction of the second author (a native of Huwwarah) with people. Milroy (2003) says: "Participants' observation can be an enormously fruitful method for sociolinguistic analysis. It precedes a tremendous supply of high quality data and crucial insight into community dynamics". Because this method faces some difficulty in analyzing the results, another instrument, the questionnaire, was needed to confirm and document more data. The questionnaire's final version included (99) words, and a five-point Likert scale was used to check the level of participants' agreement/disagreement with the use of a certain word: Strongly agree (5), Agree (4), Neutral (3), Disagree (2), Strongly Disagree (1). The evaluation relied on the following classification of the rule on the averages as follows: -Less than 2.33: Low agreement degree.
-More than the 3.67: High agreement degree.
Following the questionnaire, the study used the method of direct interview to elicit some participants' feelings about the dialect they use. The interview consisted of two questions directed to four young speakers, four middle-aged and four old people. For young people, the question was "Why don't you speak like your parents or adults in general?" and the one for old people was "What is your attitude toward your dialect?"  Postgraduate studies 3

Total 12
The following statistical treatments through statistical software packages (SPSS) version 20 for data analysis were used: -Frequencies and percentages for demographic information.
-Cronbach-alpha was calculated to extract Reliability coefficient of the questionnaire as a whole -Frequencies for "To what extent is lexical variation widespread in the study area" (n=98) -(Independent Samples t-Test) to detect the differences of lexical variation due to gender variable.
-Analysis of variance (ANOVA) to detect the differences of lexical variation due to education level and age variables.

Findings
The results of the main questionnaire show that most speakers of Irbidi dialect in the three areas of Irbid tend to abandon some of their lexical items. They also show that the most effective social variable that affects lexical variation is gender, followed by age and level of education, respectively. These variables are analyzed in relation to the use of the list of the lexical items shown below. Table 3 shows the mean of the use of each item among the whole sample regardless of gender, age and level of education. The table also exhibits the lexical items, their meanings and the level of frequency.   Table 3 shows that the highest means reached (3.85) out of (5) for lexical item (85) "dʒaxah", by high agreement degree, and then came the lexical item (28) "imTTagmiʃ", with a mean of (3.82) and high agreement degree,and then came the lexical item (34) "itʕarbaʃ", with a mean of (3.72) and high agreement degree. After that came the lexical expression (99) "indaħir min hoːn / inʕaθir", by a mean (3.67) and high agreement degree, and then came the lexical item (86) "mSoːkidʒ", with a mean (3.65)and high agreement degree, and the lowest means was (3.43) for lexical item (18) "lɑː titʃby ʕalah widʒhak", by medium agreement degree. The average mean was (3.22) for all items, representing a medium agreement degree.
In answer to the question: "To what extent is lexical variation widespread in the study area?", it was found that the participants varied in their responses depending on the specific lexical item or expression and on the social variable (gender, age, and level or education). In general, awkward or funny-sounding items were not favored by young (mean = 2.72), female (mean = 2.68) and well-educated subjects (mean = 2.92). The means (reflecting level of agreement) were higher, ranging from 3.12 to 3.91), for old, male and the least educated subjects.
Independent Samples t. Test was carried out to investigate the role of (gender) in lexical variation. The results show that males tend to stick more to older items than females who prefer newly acquired terms.
For the sake of economy and because of lack of space, Table 4 shows only the test results for the first 4 items, which are by the way representative of all lexical items under consideration. Similarly, Independent Samples t. Test was conducted to investigate the role of age in the frequency of use of the lexical items. Results indicate the people aged (55 years and above) were more keen on using traditional items than the adult age (35-54) and the young age groups (18-34 years). To detect statistical significance and explore the differences in lexical variation due to age, Analysis of variance (ANOVA) was applied; Table 5 shows results for the first 4 lexical items. The table above shows that there are statistically significant differences for the role of (age) in lexical variation. The differences due to age in all lexical items were in favor of (55 years and above), followed by middle-aged (35-54), and at last the young (18-34). Table 6 shows results for the first 4 items of the sample when we explored the effect of level of education on the use of lexical items.  The table indicates that means for (below High School) were higher than those for (Postgraduate studies education, Bachelors, and High School). To detect statistical significance and explore the differences in lexical variation due to education, analysis of variance (ANOVA) was applied. Table 7 shows results for the first 4 items.  Table 7 shows that there are significant statistical differences for the role of (education) in lexical variation. The differences due to education in all lexical items were in favor of (below High School) followed by (High School), then (Bachelor) and at last (Postgraduate Studies). To see if lexical variation in the three towns investigated affects mutual understanding, frequencies and a Chi-square test were carried out for each item to extract differences between understanding and not-understanding.  Table 8 shows results for the first 10 items: -There are statistically significant differences between Understanding and Not-understanding for most lexical items including (daʃri:ha:, go:mɑːna:, go:zaliːn, ba:Toːs, Ta:gah, ʔuxrah, tʃaʕtʃaba:n, uʕbuTiːha:, bizruTin, imTagmiʃ, itʕarbaʃ, ʔuTmuri:, inTamy, muzto:tih, yiʕta:z, ihu:ʃ ʕaleih, siddih, kadarah, fidʒʤih, daksyih, xɑːberko:, haʃuːt, birʕɑːny, bitbarTɪm, bitriːʕ /bɪtgalʕiT, goːTir min hoːn, ʃarad/ fal/mazaT, dʒaxah, mSoːkidʒ, iymuliT, indaħir min hoːn / inʕaθir) in favor of (understanding). (See Table 3 above for meanings of these items).
-There are no statistical significant differences between Understanding and Not-understanding for a few lexical items including (ɪgʕɪdiːha) in favor of (not-understanding).

The Interview
To remind, the interview contained two questions: one directed to young speakers and a second directed to older individuals. The first question was "why don't you speak like your parents or adults in general?" The goal of this question was to investigate why young speakers tend to use other lexical items instead of those of their parents. The respondents answered that they change their dialect and use other lexical items for prestige and to be different. Moreover, some speakers asserted that their aim in using certain items was to associate with their peers of the same age group.
The second question was "What is your point of view about your language?" Adults answered that they did not like to imitate others and they preferred Irbidite dialect since it is part of their identity. They also said that they were proud of speaking it. www.scholink.org/ojs/index.php/sll Studies in Linguistics and Literature Vol. 5, No. 1, 2021 108 Published by SCHOLINK INC.

Discussion and Conclusion
This study investigated lexical variation in the dialect of people living in Irbid and its correlation with age, gender, and level of education. The specific objective was to study the interaction between linguistic structure and social structure according to social variables such as age, gender, and level of education. Therefore, Labov's approach was adopted.
Three methods were used in this study to gather the information. The first was observation; a large amount of data was collected through contact with friends and relatives, who live in the three towns under investigation, and through taking notes from interactions among people. The second method was the questionnaire with close-ended questions. After collecting the lexical items by observation, a questionnaire was designed for a male and female sample of subjects belonging to different ages and levels of education. The last method was the interview; this method was used to elicit the feelings of a convenient sample of participants about the dialect they use.
According to the results in table (3), the overall mean of using the local dialect's lexical items among the sample was (3.22) out of (5)  At another level, the study results show that the variables of age, gender and education play an essential role in lexical variation. Participants aged (55 and above) have a high tendency to preserve their local dialect lexical items, followed by the middle-age group (35-45), and lastly by younger people (18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30)(31)(32)(33)(34). This finding agrees with other findings on various sorts of variation (see Al-Wer and Al-Qahtaani, 2016). Moreover, the education and gender variables, which seem interrelated in the findings, had a high effect on the results; females who are educated tend to neglect some of their local dialect's lexical items and replace them with new ones considered to be more prestigious while men use them with no reference to education level. For less educated people from the same group, the results show that subjects from both genders preserve their local dialect. The lowest mean in table (4) was for the young group (18-34), including young males and especially educated females who seem to have abandoned the lexical items used by their parents for the sake of prestige and to imitate their peers. This result agrees with Al-Essa (2006), who suggested that the highest rate of the use for the local dialect lexical items are old males and females, followed by middle-aged males and females and finally the young males. She found that young females abandoned their local dialect and used urban lexemes which were considered as being more prestigious. On the basis of the study findings, one can conclude that there is indeed lexical variation and a tendency to abandon the local tems to other in the three towns around Irbid city. Many Irbidi lexical items are being replaced by new ones among young generations especially. This tendency may subject the Irbidite dialect to a number of linguistic changes.
For future research, one may suggest investigating morphological, phonological or syntactic variation in the Irbidite or other Jordanian and Arab dialects and look at the issue from other social factors like the origin of the mother, occupation, and length of stay outside the region of the original dialect.