Operation labeling algorithm within Xamtanga sentences

: The objective of this article was to examine the operation of Labeling Algorithm within Xamtanga sentences. A descriptive research design was used to analyze the sentence structures. The data were collected from the native speakers of Xamtanga college lecturers (3 males, 1 female) and from written texts. By purposive sampling, 11sentences were selected, arranged and described. The method of data analysis working in this research was Labeling Algorithm {XP, YP}. Thus, results indicated that simple sentence structure has only one verbal head. In the case of compound, complex and compound complex sentence structures, there exists more than one verbal heads. Sentences in terms of their forms, forming syntactic object representations they contain were different. Alternatively, every sentence types share syntactic object representations that include Noun Phrase (NP), Verb Phrase (VP), Determiner Phrases (DP), Prepositional Phrase (PP), Tense Phrase (TP), Adverbial Phrase (ADVP), and Adjectival Phrase (AP). Finally, the study recommended a further research how labeling Algorithm {XP, H} and {X, Y} works to describe the label of syntactic object representations found within sentences in Xamtanga.


Introduction
Agaw is a branch of the Cushitic macro-family and part of the larger phylum generally called Afroasiatic. The Agaw people are one of the Cushitic races and the oldest ethnic groups in Ethiopia and Eritrea. They are ancient inhabitants of the Northern and Central highlands of today's Ethiopia and Eritrea. They were either forced out of their original settlements and formed a number of scattered enclaves, or assimilated with Semitic people and adopted their culture (Desalegn, 2016). The contemporary Agaw people can be classified into four categories.
Northern Agaw is Blin. They live in Eritrea, in and around Keren in Anseba zone. The Southern Agaw includes Awgni. There has been some confusion in the identification of the southern Agaw people, as this region also includes Awղi and Awi people. The ethnic group of southern Agaw is Awi live in Central Gojjam in north western Ethiopia. As a result, some publications (Tsegaye, 2013) have wrongly identified the Awi group as Awngi.
Labeling Algorithm to examine syntactic object representations found in Xamtanga sentence structures.
Syntax studies sentences and their structure, through investigating the arrangement of words and the relationship among words in a sentence. Particularly, researchers (Miller, 2008) identify how words are sequenced to build phrases, how phrases are combined to build clauses or longer phrases, and how clauses are combined to build sentences. The smallest unit of syntactic structure is a word that grammatically interacts with other such units, forming constructions on various levels. The study of syntactical structures attempts to provide set of rules that will correctly predict the possible combinations of words, which form grammatical sentences (Cinque, 2010). Researchers use visual displays of hierarchical order to describe exactly how structural relations between lexical items or words, and the way they are sequenced in a sentence contribute to the reader or listener's interpretation. Thus, the job of researchers who map syntax is to discover and formulate rules or principles that tell us how words are combined to form grammatical phrases and sentences within and across language. Generative syntax is the term to specify when the findings are reported that characterize the structure of sentences which native speakers that are grammatically correct sentences. As part of the process phrases are examined to help identify their hierarchy within a sentence, facilitating mapping of the language (Ott, 2011;Thráinsson, 2007).
A phrase is an element of structure typically containing more than one word, but lacking the subject-predicate structure usually found in a clause. It is a syntactic unit which typically consists of more than one word and is intermediate between word and clause level in sentences. In a phrase, words go together to form a single syntactic entity which can be moved 'around' and also substituted by another word (Carnie, 2013;Marques, 2011). Phrase structure is the basic unit of syntactic analysis, which is easier to see the parts of (phrases) and subparts (parts of speech) of the phrase in a tree. Apparently, syntactic tree enables to see at a glance the hierarchical structure of Phrase (Richards, 2010). Structure dependency is concerned with the hierarchical structure, normally revealed in syntactic analysis by means of tree diagrams (Pullum, 2011). Therefore, relations between any given pair of nodes contained in the same phrase marker are dominance and precedence. The relation that can be appearing between nodes in p-marker is dominance. A single node exhaustively dominates sets of nodes. Therefore, exhaustive dominance holds between a set of daughter nodes and their mother node. When the mother nodes dominate the entire set, it can be said that the mother node exhaustively dominates the rest (Sag, 2010a). Where one node contains another, the containing node is mother and the contained node is the daughter. A mother node contains several daughters, where these are said to be sisters to each other (Sag, 2010b).
Labeling Algorithm is just minimal search, presumably appropriating a third factor principle, as in agree and other operations. The relevant information about SO will be provided by a single designated element within it: a computational atom, to first approximation a lexical item LI, a head. This LI should provide the label found by LA, when the algorithm can apply. Chomsky (2013a) supposes that a label is necessary for explanation at the interfaces, and that labels are assigned by a minimal investigation, algorithm LA will be applying to [a Syntactic Object] an SO (like other operations, at the phase level). The output of the Labeling Algorithm is thus needed at the CI interface for interpretation; it licenses some Syntactic Objects (SOs) so that they can be interpreted at the interfaces. It applies at the Phase level like other operations, except External Merge that is needed to form the structure (Narita, 2011).
Syntactic object SO to be interpreted, some information is necessary about it. Thus, labeling is the process of providing that information. It is part of the procedure of forming a syntactic object SO. A fixed labeling algorithm (LA) licenses SOs so that they can be interpreted at the interfaces, operating at the phase level along with other operations (Chomsky, 2013). Chomsky (2014a) further explains that the identical labeling is required at CI and for the procedures of externalization; it must take place at the phase level, as part of the transfer operation.
The operation merge in minimalism has been taken, whether implicitly or explicitly, to include two self-determining tasks: one is to merge two syntactic objects (SOs) and the other to establish which one of the two combined SOs to project or to become the label of the resultant structure (Ishii, 2017). According to Chomsky (2013) labeling is a part of the procedure of forming a syntactic object SO. In order to examine syntactic objects in systematic way, Chomsky (2013) developed the model SO = {XP, YP}. In this model the identification of such a unique head is problematic because the structure contains two heads that are equally embedded, that the head X of XP and the head Y of YP. Consequently, LA cannot unambiguously determine which of the two heads should become the label of the structure (Blümel, 2017;Mizuguchi, 2016b;Rizzi, 2015aRizzi, , 2015bRizzi, , 2016Saito, 2016;Shim, 2018). Furthermore, Chomsky (2013Chomsky ( , 2015 Chomsky (2015) proposes to parameterize the strength of T regarding labeling: T in English-type non-natural languages, with weak agreement, is too "weak" to serve as a label, hence there must be an overt subject in SPEC-T to label the SPEC-TP as <φ, φ> by the agreeing features.
It has been widely discussed in the literature that natural language syntax and action grammar are parallel in that both involve hierarchical structures of some sort (Arbib, 2012;Knott, 2012;Moro, 2014;Pulvermüller, 2014;Stout 2010). When a Determiner Phrase and a TP are merged, they agree with each other, so that the common agreement feature can supply its label to the merged phrase, and solve the POP (Chomsky, 2013). In Xamtanga, specific morphemes unambiguously characterize nouns, verbs, adjectives, and adverbs. These are for example, more prominent in Xamtanga, with some exceptions. For instance, the morphemic suffixes added at nouns are determines and they form Determiner Phrases. The constituent marked by agreement morphology is licensed to project that agreement feature as the label of its mother node, and that agreement morphology to result from agreement with its sister. Saito's (2016) proposal about case morphology as a solution the POP, suggests that agreement and case morphology mirror each other as different solutions to the POP. If this is true, it suggests that agreement and case morphology are reflexes of each other and two different ways to resolve a POP.

Proposal
Based on Chomsky's (2013) supposition SO= {XP, YP}, neither a head. Here minimal search is ambiguous locating the heads X, Y of ZP, YP respectively. This creates the problem of Xamtanga sentence structure projections. To find solution in current research, LA defines labeling through modifying SO (by raising XP) so that there is only one visible head in the case of simple sentence structure. If, say, XP rises, then the result will be the structure with two copies of XP (Chomsky, 2013(Chomsky, , 2014Elly, 2015) as in: Then, the labeling algorithm 'sees' YP, but not XP, which is the lower part of a discontinuous element, a chain consisting of a sequence of copies headed by the structurally most important element. It is essential that a category be assigned, and the choice is stipulated to be Y=v, the verbal head of the sentence, clearly the desired outcome (Narita, 2015).
In terms of internal merge of a WH phrase, Xamtanga does not allow complementizers (C) like that, if, whatever, etc. As a result, the position of CP occupies the label of Determiner Phrase (DP). Moreover, the subject (including interrogative case) must be visible in {DP, TP} positions (Davies & Dubinsky, 2009). Sentential elements such as complementizers, sentence-final particles, aspect, tense, focus and topic, and agreement morphemes, determiners and verbs found in embedded clause are not actually the head of that phrase, which should rather taken to be silent (Leu, 2014). Moreover, Xamtanga rejects Syntactic Object movements as a syntactic operation, since they never have semantic effects (Cinque, 2014;Hartman, 2011).
Thus, I will assume, following Chomsky (2013Chomsky ( , 2013aChomsky ( , 2013bChomsky ( , 2014Chomsky ( , 2014aChomsky ( , 2014bChomsky ( , 2015; Rizzi (2016) and Shlonsky and Rizzi, 2015) is that syntactic trees must be uniformly labeled at the interfaces. Labels tell the interfaces what kind of syntactic objects they are. Hence, consistent labeling can be a consequence of interpretive principles, which may need labels to be properly interpreting structure. The other postulation that I will make use of Chomsky (2013) is that the labeler of a category created by Merge is {XP, YP} case, defined by LA that modifies SO by raising XP so that there is only one visible head Y for the entire sentence structure (Adger, 2016;Elly, 2015;Rizzi, 2015a). Y represents the main verb that is found at the end of sentence structure. On the other hand, auxiliary verbs might occur at the end of sentence structure. In this case, they correspond to T position and just help the main Verb that comes before it. T in Xamtanga is too weak to serve as a label. Throughout the analysis, XP, CP, DP, TP, VP etc, are used for expository convenience (Adger, 2016;Chomsky, 2014;Mizuguchi, 2017a Complementizers, sentence-final particles, aspect, tense; focus and topic and agreement morphemes, determiners and verbs found in embedded clause are not actually the head of that phrase in Xamtanga. Xamtanga rejects Syntactic Object movements as a syntactic operation, since they never have semantic effects (Hartman, 2011;Roberts, 2010). Likewise, Syntactic Object Representations under the tree structure is derived (i.e. formed) in a bottom-up fashion (i.e. they are built up from bottom to top).

Method
The research design used in this study was descriptive and involved gathering data which describe the syntactic object representation of Xamtanga language structural based sentences. Expert sampling was used to capture knowledge rooted in a particular form of expertise in Xamtanga. Therefore, 4 lectures in Sekota College of teachers' education (three males and one female) who teach Xamtanga were involved in computer assist telephone interview. Through this sampling based on layout, the length of data, meaning and structural simplicity as simple to display, 11 anticipated sentences were chosen for analysis. The method of data analysis employed was collection, systematic classification and description supported by the use of syntactical trees to aid the reader in understanding the structure of phrases in the Xamtanga language.

Simple Sentence
A simple sentence consists of one independent clause. Comprising a subject and a predicate, this short and independent syntactic entity intends to convey a complete idea. Xamtanga is SOV language; it may have a modifier besides a subject, object, and verb.
(1). Almaz їq'äղä dinräy aqč Almaz extremely fat became 'Almaz became extremely fat' What (1) portrays is that Almaz їq'äղä dinräy aqč is a simple sentence. The subject of the sentence is Almaz. The head of the overall sentence structure is the verb aqč. It is true that їq'äղä dinräy is the part of the VP їq'äղä dinräy aqč (Compliment of the verb aqč), it is equally true that їq'äղä dinräy is itself a phrase, namely, an Adjective Phrase (AP).

Compound sentence
Compound sentences in Xamtanga contain two or more independent clauses linked by coordinating conjunctions (like їnäղi/however). Coordination suggests that the balance of equal weight between the two clauses as in: (5). Alämu їbäzgo fu; anäղi täryäwum Alämu to the hill went up; however, didn't come 'Alämu went up to the hill; however, he didn't come' Under the analysis in (5) Alämu їbäzgo fu, anäղi täryäwum is a compound sentence which contains two independent clauses such as Alämu їbäzgo fu and anäղi täryäwum. Semicolon joins these independent clauses. The conjunctive adverb anäղi was used to join two independent clauses together. Each node under the tree has their linear and hierarchical relationship.

(6). Їjїr xasväku; Їdärä їrwuzäku
Man proposes, God disposes 'Man proposes, God disposes' Compound sentences like (6) may not require conjunction. In that case, syntactic structure of їjїr xasväku; їdärä їrwuzäku consists of S1 їjїr xasväku and S2, Їdärä їrwuzäku. In terms of relation, the mother XP node immediately dominates DP and TP daughters. Thus, DP precedes TP Node. (7). Mäkinäd mänäxaryäyil čїbrїku; anäղi їjr fiyäwum Car at bus station stopped, however no one got off 'The car stopped at the bus station; however, no one got off' In accordance with (7), the independent sentence element (represented by the XP) consists of two constituents: Sentence (S1) mäkinäd mänäxaryäyil čїbїrku and another Sentence (S2) anäղi їjr fiyäwum. Sentence S1 in turn contains the subject Determiner Phrase mäkinäd and the Verb Phrase mänäxaryäyil čїbrїku. This Verb Phrase in turn encloses Prepositional Phrase (PP) mänäxaryäyil and the Verb Phrase čїbrїku. Similarly, S2 contains bare Noun Phrase (since it was stated in S2), and the Verb Phrase anäղi їjr fiyäwum. This Phrase is further broken down into three bits: Adverb Phase їnäղi, the Noun їjr, and the Verb fiyäwum.

A Complex Sentence
A complex sentence in Xamtanga contains at least one independent clause and at least one dependent clause. Dependent clauses in the intended language can refer to the subject, the sequence or the causal elements of the independent clause.
On the subject of discussion, the present study in Xamtanga permitted the supposition it follows that every recently created SO by Merge must also contain label. The label of SO is firmed by the operation Labeling Algorithm (LA). Similar to Chomsky's postulation, the result from present study show that Syntactic Object {XP, YP}, neither a head then minimal search is uncertain, finding both the head X of XP and the head Y of YP. In order to solve this vagueness, LA defines labeling from end to end by modifying SO (by raising XP) so that there is only one visible head. However, as opposed to Chomsky (2013Chomsky ( , 2014Chomsky ( , and 2015 and Adger (2016), in the cases of complex, compound and compound complex sentences, there exists at list two verbal heads. Within these differences, then the Labeling Algorithm looks YP, which is the lower part of a discontinuous constituent, a sequence consists of a chain of copies headed by structurally most significant element.
Comparable to Shlonsky and Luigi (2015) study, the principal hypothesis in the present study was that syntactic trees were consistently labeled at the interfaces. Thus, regular labeling can be a product of interpretive principles, which may require labels to be properly interpreting structure. The next hypothesis that the present research used Chomsky's study (2015) was that, the labeler of a group created by Merge was {XP, YP} case, defined by LA that modifies SO by raising XP. The major distinction between this study and the above research works was that, sentential elements such as complementizers, sentence-final particles, aspect, tense, focuses, topic and agreement morphemes, and determiners in Xamtanga are not actually the head of that phrase. Like Cinque's (2014) and Hartman's (2011) findings, Xamtanga rejects Syntactic Object movements as a syntactic operation, since they never have semantic effects.

Conclusion
The model [XP, YP] structure is problematic for minimalist Labeling Algorithms, which rely on structural asymmetry to identify the label (Chomsky, 2013(Chomsky, , 2014Elly 2015). At this point, minimal search is ambiguous locating the heads X, Y of ZP, YP in the same way.
To determine decision, LA defines labeling all the way through modifying Syntactic Object (by raising XP) so that there is at least one noticeable verbal head in a simple sentence structure. Afterward the Labeling Algorithm 'sees' YP, but not XP, which is the lower division of a alternating constituent, a succession consisting of a sequence of copies headed by the structurally most important element. It is essential that a category be assigned, and the choice is stipulated to be Y=v, the verbal head of the sentence, clearly the desired outcome in Xamtanga Syntactic Object representation (Narita, 2015). Syntactic Object representations found in sentence structures include: Noun Phrase (NP), Verb Phrase (VP), Determiner Phrase (DP), Prepositional phrase (PP), Tense Phrase (TP), Adverbial Phrase (ADVP), and Adjectival Phrase (AP).