The Geography of Dayak Dialect in Landak Regency, West Kalimantan

The aims of this research entitled “The Geography of Dayak Dialect in Landak Regency, West Kalimantan” were to describe lexical variation, calculate the difference in lexical variation between observation points, map lexical variation, and create isogloss documents lexically. The form of this research was descriptive qualitative; the source of data for the Dayak language used in Landak was the native speakers of Dayak language. The methods used for data analysis were synchronous comparative method and triangular comparative technique between villages; while the formula used to analyze the data was the Dialectometric. The results from the data analysis, difference in lexical variation calculation between observation of Dayak language in Landak regency indicated that there found language differences, dialect differences, sub-dialect differences speech differences and no differences. Different lexical variations existed in the study area. There found 2 languages from the mapping of lexical variations of the Dayak language in Landak Regency. The first language included 4 observation points (Air Besar/Serimbu) called as S language (Serimbu) and the second language at observation points ((1, 2, 3, 5, 6, 7 / Sompak, Darit, Landak, Pahoman, Sengah Temila, and Mandor ) is referred to as the K language (Kanayatn). The K (Kanayatn) languge itself has 2 dialects, namely first dialect at the Observation point (3) Landak (members of Landak), the second dialect at the point of observation (1,2,5,6,7) Darit dialect (members of Sompak, Darit, Pahoman, Sengah Temila, and darit). The lexical isogloss documents separated the different language


INTRODUCTION
The study of dialect geography or language mapping is an interesting study in language researches. Dayak language is one of the many languages found in Indonesia, especially in West Kalimantan. The first step in the study of the Dayak language needs to be stated about the origin of the speakers and the origin of the language which are now known as the Dayak language. Several linguistic figures and archaeologists have expressed their opinion on the origin of the Dayak speakers. Dyen (cited in Fernandez, 2004) believed that the Austronesian languages originated from Papua New Guinea, New Hybrid, or Formosa. This was based on linguist studies using lexicostatistical techniques, employing 352 Austronesian lexicons.
Bellwood (cited in Fernandez, 2004) stated that the land of origin of Austronesian speakers was Taiwan (Formosa). Chang's reason for determining Formosa (Taiwan) as the origin of Austronesian speakers was based on the similar artefact discoveries found in Taiwan and in the Indonesian archipelago. Bellwood and Blust (cited in Tanudirjo & Prasetyo, 2004) described based on archaeological evidence, the earliest Austronesian settlements were between 4000 to 3000 BC. Archaeological evidence was used due to the finding of pottery culture in Taiwan. These cultural and linguistic features existed in Taiwan for about 1000 years before appearing on the southern island of Taiwan. However, the emergence of Proto Austronesia (PAN) did not coincide with the period of early Austronesian settlement in Taiwan. The early Austronesian communities in Taiwan probably spoke one or more languages which eventually formed two language subgroups and the two have survived separately until now. The separation of the Malay Polynesian (MP), Proto Austronesian (PAN) and Formosan subgroups is estimated to be 3000 BC, the separation of the PMP subgroups lasted 2500 BC, the separation after PMP in the Philippines was around 2500 BC, while the migration to Kalimantan was around 2000 BC. The distribution of Austronesian speakers originating from Taiwan (Formosa) is known as the Out of Taiwan model.
The Dayak language is a family of Austronesian languages. The Austronesian language family is divided into several branches and sub-branches. The following is the classification of the Robert Blust language that has been accepted by most Austronesian language researchers. Proto Austronesian is divided into: (1) Formosan language, (2) Polynesian Malay. Polynesian Malay is divided into: (1) West Polynesian Malay, (2) Middle-East Polynesian Malay. The Dayak language is one of the languages belonging to the West Polynesian Malay family.
There is no specific clarification or explanation of the origin of the language related to the opinions of the origin of Dayak language which is now used by Dayak speakers on the island of Borneo. Previous research on the geography of the Dayak language dialect, Patriantoro (2018) examined "Geography of Dayak dialects in Bengkayang Regency, West Kalimantan". The results revealed that first this study describes the variation of the Dayak language by lexical mapping. Second, the diachronic reconstruction has not been carried out. Third, based on the descriptions of the research in Bengkayang Regency, three Dayak languages were found. It was believed that each Dayak language has different language variations. Moreover, the research merely discussed the synchronous research to map the Dayak language in Bengkayang Regency.
The Dayak language in Landak Regency has different language variations in every district. Yet, the existing language variations cannot be ascertained as variations in language with different accents or speech, different sub-dialects, different dialects, and possibly different languages. To find out the variations in the Dayak language in Landak Regency, a dialect geography research is needed. The Dayak language used in this area, based on the speakers of the Dayak language, is called the Dayak Kanayatn language. The area of distribution is in the hinterland of Landak Regency. Collin & Shin (2008) calls it as the Dayak Gunung. Lynch (1998) explained the conditions of language on the island of Kalimantan, especially language mapping in West Kalimantan. The distribution of languages in West Kalimantan includes: (1) Malayan Group which consists of (Malay Subgroup, Malayic Dayak Subgroup, Iban-Isolate); (2) Land Dayak; (3) Mbaloh Group. Geographically, Malay speakers in West Kalimantan spread along the coast in Sambas, Bengkayang, Pontianak, Kubu Raya, Ketapang, and North Kayong districts, downstream of the Kapuas river, entering Pontianak city slightly upstream of the Kapuas river. The area of Malay speakers is called the Malay Subgroup.
The mapping of Lynch (1998) in West Kalimantan showed that the distribution of the Dayak language was far from the coastline. This language was spoken in the headwaters and mountainous areas. The distribution of the Dayak language, as well as the Dayak speakers was partly spread in the mountains and upstream of the rivers in West Kalimantan. The results of the mapping stated that the upstream area of the Landak river was the area where the Dayak language 'Land Dayak' was used (Lynch, 1998). Likewise, the areas away from Landak River stream were where the Dayak language 'Land Dayak' was spoken. Based on linguistic facts, the distribution of the Dayak language in Landak Regency is now using the Dayak language 'Land Dayak', especially in the upstream area of the Landak river and areas that are some distance from the Landak river. The fact that is emerging now is that in every village, subdistrict in Landak district, Dayak speakers are confirmed to be existed. In 2020, the Dayak language distribution will reach out to the sub-district and districts in Landak Regency. Alloy et al. (2008, pp.42-113)  From the aforementioned statements, this researcher of this research entitled "Geography of Dayak Language Dialect in Landak Regency, West Kalimantan" proposed 4 objectives, namely: (1) to describe the lexical variations of the Dayak language in Landak Regency; (2) to calculate differences in lexical variations between observations of the Dayak language in Landak Regency; (3) to map the lexical variation of the Dayak language in Landak Regency; and (4) to create lexical isogloss files in Dayak language in Landak Regency.

LITERATURE REVIEW
Dialect geography is another name for dialectology. In subsequent developments, dialectology focuses more on the study of dialects in a language. Dialect geography studies language variations based on local differences (places) in one language area (Laksono & Savitri, 2009). These language variants can arise due to geographical differences (Ayatrohaedi, 1979, p.6). Dialect geography is an attempt at dialect mapping. Language variations that are not yet known with certainty include language, dialect, sub-dialect, and speech differences are referred to as isolect (Mahsun, 2010, p.46).
The term dialect according to Meillet (1970, pp. 69-71) is usually based on variations of the same language used in different geographic areas. In general, dialectology is the study of certain dialects or dialects of a language (Laksono, 2004). According to the view of dialectologists, all dialects in a language have an equal position, the same status, there are no dialects that are prestigious and not prestigious (Laksono & Savitri, 2009). In its principle, each dialect of the same language has the same role and function as a means of communication in community groups to establish social relationships with others.
Maps are representations through images of an area that state boundaries, surface properties, latitude, soil structure, and natural conditions. Language mapping means transferring the language data collected from the study area onto the map. Laksono & Savitri (2009) stated that there are three types of maps in dialect geography research, namely: (1) base maps, (2) observation point maps, and (3) data maps. First, the base map is a geographic map related to the study area. It has to be displayed to determine the administrative boundary observation points. Second, the map of the observation points which contains the area of observation points for which data is taken. The name of the observation point is written in numbers and the name of the observation point is written in the description. Third, the data map contains research data at each observation point. Some research data are placed directly at each point of observation, and some use symbols.
Isogloss is an imaginary line that unites regions of the same language variation (Lauder & Lauder, 2009, p.221). The similar definition is stated by Keraf (1984, p. 54) that isogloss is an imaginary line that connects each observation point that displays similar linguistic symptoms. The word isogloss comes from the word iso + glos. Iso means 'equal / not diverse', and glos means 'smooth surface'. Isogloss is an imaginary line connecting each observation area that displays similar linguistic symptoms, then the concept develops into an imaginary line that unites the regions of observations displaying similar linguistic symptoms (Laksono & Savitri, 2009, p.91). Kurath (cited in Laksono, 2004) states that heterogloss is an imaginary line inscribed on the language map to separate the emergence of each language symptom based on different forms or systems. Each creation of an isogloss file is carried out by assigning certain symbols to each number. Berian which has similar linguistic symptoms uses the same symbol. The combination of berian has the same symbol as the isogloss line. The line can be curved or straight and drawn between the DP.
The display map is a map that contains tabulations of field data with the intention that the data is drawn from a geographical perspective. Thus, the demonstration map includes the geographic distribution of differences in linguistic elements between the observation areas (Mahsun, 2005). If one examines the phonological and lexical differences, then all the berian which has phonological and lexical differences are mapped into two different display maps (Laksono & Savitri, 2009). Data with phonological differences are mapped in a phonological display map. Data with lexical differences are mapped in the lexical display map. (Ayatrohaedi, 1979) mentioned 3 maps of creation including: (1) direct system (moving each number onto the map), (2) symbol system (replacing the berian with certain symbols), and (3) plot system.
The lexicon is the technical term for the language component. Verhaar (2008) stated that the term lexicon in linguistics means vocabulary which is often called "lexeme". In line with this opinion, Kridalaksana (2009) stated that the term lexic is used in British linguistics. The popular term is vocabulary that has the same meaning as the two terms. Subroto (2011) claimed that lexemes are essentially abstract forms or the results of abstraction of different word forms that are included in the same lexeme contained in the same paradigm which is called the inflectional paradigm. Therefore, the lexeme is an abstract unit (the result of an abstraction) of an inflectional paradigm that does not change the identity of a word or class of words as the smallest form, either simple or complex.
This research employed quantitative research and qualitative research. These two types of research were used to complement each other in data analysis. Some of the data can only be analyzed quantitatively; and some must be analyzed qualitatively. Quantitative research is a research that uses numerical counts using certain measurements. Measurement is an activity involving giving numbers to attributes, characteristics of a person, objects, or events according to rules or formulas. Measurement is the process of assigning numbers to certain categories to describe the quality of certain results. Measurement in dialect geography uses the formula of "Dialectometric".
A qualitative research methodology is a research that does not simply use numbers, but an approach that describes the actual situation to support data presentation. The researcher analyses the data in all its richness of character as closely as possible to its original form as at the time it was recorded. In addition, it supports the use of 'Tacit Knowledge' which is intuitive and perceived as additional knowledge that can be expressed in the form of language. In collecting data, qualitative researchers do not only record what is formally stated, but also the things they feel and perceive intuitively (Sutopo, 2002) This type of quantitative research was used in this dialect geography study. Quantitative research was used to calculate lexical differences between observation points. The formula used was the formula of "Dialectometric". The result of calculating the lexical difference between observation points used triangles between villages found lexical language variations. The results of calculating the linguistic distance as a percentage in the field were found to be different languages, different dialects, different speeches, and there was no difference, if the difference was less than 20.1%. Specifically, for lexical calculations, Synchronic Comparative was used with the non-cognate pair technique to find lexical differences which included lexical pairs with different shapes and different or the same meaning.
Sources of data in this study were native speakers and dialogues. The data source for native speakers was native Dayak speakers who were born, lived, and raised in Landak Regency. The source of the dialogue data or the informant's conversation that the researcher had planned was in accordance with the instrument and the time that had been prepared and determined.
The data of this research were in the form of the Dayak language words used in Landak Regency whose gloss was already determined. In data collection, the context of the data was also important. The function of the data context was to familiarize the data or lingual units, as well as to check their accuracy through other data contexts. Sudaryanto (1988) stated that the context of the data is the basic component that became a requirement for data. The glossary in question was in the form of Swadesh lexical and other lexical that were not Swadesh, with 100 lexical in total. The instrument used was a modified Nothofer instrument by Laksono & Savitri (2009, pp.45-60) The data collection method used in this study was a proficient method, because it was a conversation and there was a contact between the researcher and the speakers as the resourceful persons. The researcher was intended to obtain data with his ingenuity and ability to lure informants through the research instrument guide so that the informant could speak as expected by the researcher. This fishing technique was carried out by direct "face to face" conversation. The conversation was controlled by the researcher and directed according to the researcher's interest to obtain data as complete as possible, as much as the intended type of data (Sudaryanto, 1988, pp.7-8).
The instrument used in the data collection was the Nothofer instrument modified by Laksono & Savitri (2009). The instrument was in the form of Swadesh words and other words with a total of 100 glossaries. The instruments, then, were written in glossy form. Laksono & Savitri (2009, p.23) stated that gloss was a form known in the language used by researchers. Therefore, the research instrument used in the glossary was the equivalent form in Indonesian. Sudaryanto (1988, pp.26-32) stated that informants or language assistants were resource persons for research materials, information givers, and research assistants in obtaining data. Informants were selected with certain criteria. The number of informants in the study "Geography of the Dayak Language Dialect in Landak Regency, West Kalimantan" was 3 people for each observation point. Since the research was conducted in 7 points of observation, so there were 21 informants in total.
The methods used in data analysis included synchronic comparative methods and diachronic comparative methods. Sudaryanto (1993) used the term equivalent method to express a comparative method. Matching method was a way of analyzing language data by matching or "comparing" one data with another. Crowley & Bowern (2010) stated that the comparative method was a way of working to "compare" two or more cognates from two or more languages in order to get the proto-language form. The comparative method can be used to reconstruct several aspects of the original language from the reflection of the mother tongue, only the reconstructed language data were those which were 'cognate' related.
The method used for analysis, especially language mapping, was the synchronous comparative method. In its principle, the synchronic comparative method was a method used to analyze the language data by comparing the language data with observation points in the same period of time. The diachronic comparative method was a method used to analyze data by comparing language data for different time periods. In this study, the diachronic comparative method was not used. The next step was calculating the number of lexical differences amongst observation points. The number of lexical differences between observation points was known. The results of the lexical differences were calculated using the "dialectometric" method. Dialectometric was a statistical measure used to see how far the differences were in the places studied by comparing a number of elements collected from certain places (Laksono & Savitri, 2009). Guiter Dialectometry Formulas (cited in Mahsun, 2005Mahsun, , 2010 can be seen below: (S X 100) = d % n S : total of lexical differences among observation points n : total of lexical maps to be compared. d% : Percentage of distance between linguistic elements between observation points.

FINDINGS AND DISCUSSION
Based on the research results of Alloy et al. (2008, pp.42-113) using ethnolinguistic research, there found 17 Dayak languages in Landak Regency. Ethnolinguistic research was linguistic research in which the determination of language or language variations in a language community area was determined by the language-speaking community itself. So, it is normal if the research results of Alloy et al. (2008) in Landak District revealed 17 Dayak languages.
This research certainly produces language and language variations that are different from those done by Alloy et al. (2008). This research uses dialectical geography or dialectology research. Determination of language variations or different languages are based on the calculation of lexical differences between observation points using inter-village triangles with the Dialectometric formula. The results of the discussion of data analysis "Geography of Dayak Dialects in Landak Regency, West Kalimantan" are presented below.
The results of data analysis are based on (1) lexical data descriptions; (2) calculating the lexical variation difference between observation points; (3) lexical language variation map making, and (4) lexical isogloss file map making. The analysis of four research problems indicated the following results.
First, the data used for the lexical variation analysis were 100. The data were written phonetically. The calculation of lexical difference phonological difference was not taken into account. In the analysis of lexical differences between points, the overall observation was carried out for words which included Swadesh and not Swadesh. Furthermore, the calculation of the lexical difference between the points of observation as a whole can be seen in table 1 below. (1) the difference between the observation points 1-2 = 20, 5 -6 = 19.5 and 6-7 = 19.5; (2) the difference between observation points 1 -6 = 24, 1 -7 = 21.5, 2 -5 = 23, 5 -6 = 29.5; (3) differences between observation points 2 -6 = 31.5, (4) differences between observation points 2 -3 = 65 and 3 -5 = 67; and differences between observation points 1 -4 = 86, 2 -4 = 82.5, and 3 -4 = 86. Based on the facts in the field, differences in language variations between observation points are not the same. In terms of language variations, for example, the Dayak language, even though the language is used in the same district, is used in different villages and sub-districts as a result of differences between the points of observation for each of the numbers might be different or the same. Languages that are considered the same in the same district are not necessarily the same language. As observation points 1 -4, 2 -4, and 3 -4 the difference is more than 80.1. The observation points 5 -7 and 6 -7 also have the same number of lexical differences as 19.5.
Second, the calculation of the lexical difference between observation points using the Dialectometric formula resulting in "Linguistic Distance in Percentage" is show in Table 2 below. Based on the calculation of the overall lexical difference and the calculation of the percentage of lexical distances between points using the dialectometric formula, it can be seen that there is no difference between observation points 1 -2 = 20%, 5 -6 = 19.5% 6-7 = 19.5%. Speech differences are found between observation points 1 -6 = 24%, 1 -7 = 21.5%, 2 -5 = 23%, 5 -6 = 29.5%. Sub-dialect differences are found to be 2 -6 = 31.5%. Dialect differences are found between observation points 2 -3 = 65% and 3 -5 = 67%. Language differences are found between observation points 1 -4 = 86%, 2 -4 = 8.52%, and 3 -4 = 86%. There are 2 Dayak languages found in the research area in Landak Regency. Furthermore, the calculation of the overall lexical difference between observation points, the linguistic distance between observation points in percentage can be seen on map 1 below. Third, lexical mapping using multiple facets is carried out to determine the exact and precise areas of the Dayak language variation, lexically can be seen on the map below.

Map 2 Lexical Dialectometric Polygon Map
The Map 2 lexical dialectometric polygon map above is interpreted that in Landak Regency, there are 2 language groups of the Air Besar (4) Language S and the Landak group consisting of (Sompak, Darit, Landak, Pahoman, Sengah Semila, and Mandor/1, 2 , 3, 5, 6, 7), and K language. The K language consists of two dialects; The difference between observation points is said to be language difference, if the lexical difference between observation points is 80.1% and above and the difference is below 0% -20%, it is considered not a difference.
Fourth, the creation of an isogloss file based on the calculation result of "Linguistic Distance in Percentage" produces an isogloss file. There are two languages in the Dayak language in Landak Regency. The difference between observation points is said to be different in language, if the lexical difference between observation points is at least 80.1% (Patriantoro, 2015). The difference between observation points 1 -2 = 86%, 2 -4 = 82%, and 3 -4 = 86%, the line is a language dividing isogloss file, which limits the S / 4 language (Serimbu) with K / 1, 2, 3, 5, 6, 7 (Kanayatn). The line is a dialect dividing isogloss file between the Landak dialect (3) and the Darit dialect (1, 2, 5, 6, 7). The difference between other observation points, the difference in speech between 20.1% -30% and the difference below 0-20% is considered not a difference.