ELLIPSIS AS A COHESIVE TIE

: In this paper we have tried to analyze the cohesive aspect of elliptical constructions that people use in language in order to communicate with each other fluently and without interference. In the light of this, we described the forms of cohesion, ellipsis, and the situational integration of these phenomena within the literary genre, and the ways in which their contents ensure textuality. Special attention is paid to the dominant form of these connections in literature, which are explained in detail through illustrative examples taken from such texts. We have focused on problems and issues related to the textual reference and the general notion of cohesion, coherence and ellipsis.


Basic Notions -Text/Discourse and Cohesion
Discourse analysis is most commonly defined as an analysis of language in use. As such it cannot be limited to the description of linguistic forms independent of their purposes or functions, which are actually designed to serve people in their functions. Thus, discourse analysts are dedicated to researching what that language is actually used for. While the formal approach has a long tradition, the functional form is somewhat less documented. It is unlikely that any natural language actually has only one function to fulfill on a given occasion, and that another is completely excluded.
An important concern of discourse analysts has always been to study how the use of a given discourse contributes to cohesion and coherence. Cohesion and coherence are semantic concepts: cohesion has to do with relations between text and syntax, and coherence has to do with knowledge or cognitive structures, which are implied by the language used and which contribute to the overall meaning of a given discourse. Brown and Yule (1983) describe that function in which language serves to express 'content' as transactional, and that function of language involved in expressing social relations and personal attitudes is described as interactional. Linguists and linguistic philosophers tend i Correspondence: email vera.vujevic.djuric@ff.ues.rs.ba to adopt a limited approach to the functions of language in society. Although they often admit that language can be used to perform many communicative functions, they nevertheless make a general assumption that the most important function of language is communication.
The value of using language to convey information is well embedded in our cultural mythology. We all believe that our ability to speak allowed the human race to develop into different cultures, each with its own distinctive social and religious customs, laws, oral traditions and folklore. Moreover, we all believe that it was the acquisition of a written language that enabled the development within some of these cultures of philosophy, science, and literature. Thanks to this ability to transfer information through the use of language, people were able to use the knowledge of their ancestors, as well as the knowledge of other peoples and other cultures.
The language used to convey factual information will be called primarily transactional language. In the transactional function of language, we assume that what the speaker (or writer) primarily has in mind is the effective transmission of information. The language used in such a situation is primarily message-oriented. It is important that the recipient gets the detailed information correctly. Thus, if a policeman is giving instructions to a passenger, if a teacher is explaining students the new lesson, if a doctor is telling a patient how to use medicine, if an aggrieved customer files a complaint, a salesperson explains to a customer the exact specifications of a product, or a scientist describes an experiment, in each case it is important that the speaker should clarify what is said (or what is written). If the recipient has not understood the message correctly, the consequences in the real world can sometimes be even catastrophic.
Sociologists and sociolinguists are particularly concerned with the aspect of using language to establish and maintain social relationships. It is clear that a large part of everyday human interaction is characterized not primarily by interactional but by transactional use of language. Therefore, conversation analysts are primarily concerned with researching the use of language used for negotiation, and the role of the relations between interlocutors. There is also peer solidarity, exchange of turns and gestures in conversation, facial expressions of both the speaker and the listener. When two strangers are standing in a queue at the supermarket and one turns to the other and says "My God, prices nowadays are awful", it is difficult to assume that the speaker's primary intention is to convey information. It seems much more reasonable to suggest that the speaker is showing a willingness to be friendly and talk. Indeed, a large part of ordinary everyday conversation seems to consist of one person commenting on something that is close to both him and his interlocutor. The weather is, of course, the most quoted example of this in British English, although it has carried over into most cultures and languages. Halliday and Hasan (1976) believe that the primary determinant of whether a set of sentences constitutes or does not constitute a text depends on the cohesive relationships within and between sentences, which create texture. Texture is provided by a cohesive relationship. These authors also offer taxonomy of the types of cohesive relationships that can be formally established within a text, providing cohesive links that bind the text together.
The concept of texture is entirely appropariate to express the property of "being a text". A text has texture, and this is what distinguishes it from something that is not a text. It derives this texture from the fact that it functions as a unity with respect to its environment (Halliday and Hasan 1976: 2) Glovacki- Bernardi (2001: 236) states that the most important concept in the field of text linguistics is actually coherence. She states that it means "grammatical, syntacticsemantic relations between sentences and other linguistic elements in the sequence of sentences." Cohesive relationships within a given text are established where the interpretation of an element in the discourse depends on the interpretation of another element. One presupposes the other in the sense that it cannot be effectively decoded unless one resorts to the interpretation of the first element. A well-known type of explicitly marked cohesive relationship within texts is indicated by formal markers that connect what will be said with what was said before -markers such as and, but, so and then/then. Of course, it is not the case that any of these formal markers stand in a simple one-to-one relationship with a particular cohesive relationship: and, for example, they may appear between sentences showing any of the conjunctions -causative, temporal, or adversative. This is best illustrated by the following example: Example 1: "We ended up going for a drink and then a meal in a Bernie's Inn. Returned chez Jane for coffee and talk. Bed about midnight." (Brown and Yule 1983: 192) Although the sequential nature of events is only explicitly indicated by the markers then between going for a drink and a meal in a Bernie's Inn it is clearly implied, although not stated, in the sequence of events that follow. Basically, we can conclude that it is actually the fundamental semantic relationship that has the power of cohesion, and not a certain cohesive marker. Yet the presence of cohesive markers is what constitutes a text.
Also, there is no single or unique unit that can be used for all types of discourse analysis. In general, we can say that scientists rejected the sentence as a unit, because their research aims to go beyond the sentence, so one of the initial (formal) definitions referred to a language unit larger than a clause or a sentence (Alba-Juez, 2009: 291). But the phenomenon of discourse is more embedded in the functional definition of language ("language in use") than in the formal one, and the functional definition says nothing about the size of the units. Depending on the level at which they focus, researchers can deal with larger or smaller units. Different approaches work with different units, but the same analyst can handle different units simultaneously if they consider it appropriate for the purposes of their study. So, for example, for a sociolinguist's interaction, the main unit of analysis may be an utterance or strategy of politeness, while for a discourse analyst it will be a much broader and larger unit than an utterance or a strategy. The term discourse is generally applied to interactions that can be very long, medium length, or very short and simple like 'Hello!' during an introduction:  (SND,52) Most linguists support the view of discourse as linguistic interaction covering any length. And if discourse can span any length, then the units selected for analysis will also be larger or smaller, depending, among other things, on the type of discourse used as a corpus. Depending on the perspective of the analyst or the variables considered, there are different types of discourse, such as legal discourse, medical discourse, scientific discourse, computer-mediated discourse, or family discourse. Alba-Juez (2009: 292) gives the following taxonomy of the relationship we see between the forms used and the concrete discourse: formal and informal, reciprocal/non-reciprocal, spontaneous/unspontaneous, face-to-face telephone conversation, public/private, taskoriented (discourse oriented towards for a specific purpose (e.g. psychotherapy), written (includes all forms of linguistic communication in writing), memorable (intended to last, to continue, to be recorded for the future), empathic (we can see what each participant feels, e.g.: a normal face-to-face conversation face-to-face, dialogue), speech-visual (gestures, movements, etc.) to name a few.
However, it should be noted that no discourse belongs to a unique and exclusive type. There are no absolute differences between all the different types of discourse, and therefore we can more properly speak of a continuum types of discourse rather than separate and distinct categories. For example, a conversation between a professor and a student at the end of class may be located somewhere between the formal/informal range: there is a certain level of formality due to the distance and power difference between the student and the professor, but at the same time the specific situation does not require a high level of formality, so the analyst will certainly find certain characteristics of informal speech in this conversation as well.

Ellipsis and Cohesion
Cohesive connection is most often realized through the following linguistic mechanisms: reference, conjunctions, substitution, ellipsis and lexical cohesion (Halliday andHasan 1976: 4, Brown &Yule 1983: 192). The term ellipsis is used to denote that cohesive connection in which one part of the structure is omitted, and can be reconstructed based on the examination of the adjacent context. We can look for the interpretation of omitted material outside the text itself, in the context of the situation, and this relationship is said to be an exophoric relationship, which plays no role in textual cohesion. Where the interpretation lies within the text, they are called endophoric reference and form a cohesive link within the text. Endophoric reference is divided into two types: anaphora and cataphora. Anaphoric words refer to what was previously said, and they look back in the text for their interpretation, while cataphoric reference relies on some other linguistic unit that follows it for its interpretation. Brown and Yule (1983: 193) illustrate this very simply as it follows: This question, how to distinguish a text from a non-text, leads us first to different historical approaches to the phenomenon of text. As stated by Glovacki-Bernardi (2000: 236) "at the beginning of text linguistics there is first the research of super-clausal referring elements, which include pronouns, super-clausal particles with anaphoric function (referring backwards) and cataphoric function (referring forwards), as well as various connectors." This type of text analysis is primarily directed towards the linguistic composition that expands the linguistic area of the sign "from the area of the linguistic level of the sentence to the text" (2000: 236). This is shown by the following examples that we selected from the corpuswe have an anaphoric reference (example 4) where we draw the interpretation from the antecedent that comes before the pronouns she/her. Examples (5, 6) is a cataphoric reference: Examples 4, 5, 6: "Blanche comes around the corner, carrying a valise. She looks at a slip of paper, then at the building, then again at the slip and again at the building. Her expression is one of shocked disbelief." (SND, 11) "We have a name for it. We call it 'the bag'." (HWSG, 4) "It's kind of cute, the IV bag." (HWSG,4) Studying elliptical constructions through the literature, we have noticed that there are three basic types of ellipsis: 1. Syntactic ellipsis -indicates the incompleteness of the surface structure. This type of ellipsis includes examples in which obligatory linguistic elements for syntactically complete sentences are omitted (predicates, for example), but under strict conditions related to form and consistency of interpretation. 2. Semantic ellipsis -constitutes the omission of syntactic elements that are not mandatory (modifiers in noun phrases, for example), but are required to complete the semantics of the text interpretation. 3. Pragmatic ellipsis -speaks of the incompleteness of the interpretation, which is determined according to the general principles of pragmatic interpretation. The largest number of studies of the phenomenon of ellipsis uses the theoretical framework that Halliday and Hasan (1976) stated in their research: nominal ellipsis, verbal ellipsis, clausal ellipsis.
In our paper, we will combine the two approaches, because it is necessary to explain the ellipsis as a cohesive connection. Nominal ellipsis, as one type with very strong context-dependent cohesive tie, will help us to exemplify this connection between ellipsis and cohesion.
In order to distinguish ellipsis from other linguistic means of omission, it is very important to point out that the principle of verbatim recoverability is applied to ellipsis, which implies that the actual deleted word(s) must be precisely recoverable. Since it is very difficult to set clear boundaries of elliptical constructions, the happiest solution is to establish different degrees of strength in identifying examples of ellipsis.
Let us first consider the examples that we selected to illustrate the rules formulated in this way: Each example meets the stated criteria. The first mentioned criterion, which is followed by all the others, implies that in such context where there is no referential ambiguity, it is clearly determined which words should be reconstructed. The elided material can be supplemented and interpreted based on the context, i.e. adjacent text: This is my house and I'll talk as much as I want to talk! (example 7). After deletion, usually of identical structure and already known information, the elliptical sentences really remain structurally truncated and defective: I thought she would be/I feared she would be because these elliptical sentences would be ambiguous without proper contextual reference.
Although some structures are clearly defective, they do not meet the criteria of being accurately reconstructed. Although they behave like complete statements, in most cases they lack certain grammatical structures that are very often impossible to analyze according to some established grammatical pattern. Examples of the criterion according to which elliptical constructions are grammatically defective, without the possibility of referring to a linguistic or situational context, are the statements Hello and Thanks. These statements in the first-place show that they were not eliminated according to the first criterion we mentioned, as well as that this criterion of defectiveness of elliptical constructions cannot be usefully applied without the criterion of literal reconstruction. We must keep in mind that what guarantees ellipsis is the possibility of its reconstruction from the adjacent text.
Here, the concepts of coherence and cohesion in the interpretation of the text really play a key role, because they are useful in creating connections and relationships between sentences, statements and communicative events. What is necessary in achieving such a goal is to look at the purpose of each individual element and its contribution to the functioning of the text as a whole. Each linguistic unit, in fact, has its own specific role and distinctive feature, and as such affects the stability of the text. De Beaugrande and Dressler (1981: 60) describe in a very simple and clear way that 'the stability of the text as a system is maintained by the continuity of events.' What is pragmatically needed in analyzing such examples is a term that would refer to one instance of cohesion, a term for one representation of cohesively connected items. We will call this phenomenon a connector or a tie, as defined by Halliday and Hasan (1976: 3). The relationship between the phrases he and some small person in example (9) is actually what makes a cohesive tie in a text: Example 9: "At this point, some small person tugging on my sleeve informed me that he wanted a strawberry milkshake, not a chocolate one." (RLD,39) In addition, example (9) contains a form of substitution between the noun phrase a strawberry milkshake and a chocolate one. The concept of this connection enables the analysis of the cohesive properties of a text and provides a systematic representation of its textural patterns. In this way, different kinds of phenomena in language can be investigated, for example, the differences between speaking and writing, about the relationship between cohesion and the organization of written texts into sentences and paragraphs; about possible differences between different genres and different styles of writers and the kinds of cohesive connections they usually use.
We see that cohesion occurs where the interpretation of some elements in the discourse is dependent on the interpretation of others. One presupposes the other, in the sense that it cannot be effectively decoded unless we return to it and reconstruct the meaning. When this happens, a relation of cohesion is established, and the two elements, the presupposition and what it presupposes, are at least potentially integrated into one text. Cohesion is part of the system of every natural language. The potential for cohesion lies in the use of ellipsis, which as universal language mechanisms is built into language itself. The actualization of cohesion in any given case, however, depends not only on the selection of some option from these resources, but also on the presence of some other element that resolves the assumption this poses. Like any other semantic relations, cohesion is expressed through the layered organization of language. Language can be explained as a multiple coding system consisting of three levels of coding: semantic (meaning), lexico-grammatical (forms) and phonological and orthographic (expressions). Meanings are realized (encoded) as forms, and forms in turn are realized (recoded) as expressions (Halliday and Hasan 1976: 5).
The guiding principle in language is that more general meanings are expressed through grammar, and more specific ones through vocabulary. Cohesive links fit into the same general pattern. Cohesion is expressed partly through grammar and partly through vocabulary, so we can talk about grammatical cohesion and lexical cohesion. Reference, substitution and ellipsis are grammatical, which means that their cohesion is realized through the grammatical system, as in the following elliptical example: The second sentence of example (10) is coherent through ellipsis, where I know / But I wanted to are the presuppositions of the sentence with the full meaning I know I didn't have to do that / But I wanted to do that. And, as we have already mentioned, in this part of the our paper we will analyze nominal ellipsis in English as just one example of contextdependent ellipsis, and one of the most common cohesive links within a certain discourse.

Context-Dependent Ellipsis 2.1.1 Nominal Ellipsis
In this part of the paper a cohesive connection within context-dependent elliptical constructions is presented through just one example of textual ellipsis -ellipsis of a noun phrase, i.e. nominal ellipsis. Namely, ellipsis is defined as a complex phenomenon. In linguistics, it means the omission of linguistic elements that should be understood from the context, where the recipient should adequately fill the grammatically permitted gap. Elliptical noun phrases are cohesive and usually refer anaphorically to another noun phrase, but the main question is how much of the full noun phrase is included within the elided structure. For this purpose, we will analyze the examples to define which elements of the noun phrase can be omitted and under what conditions in nominal ellipsis.
Cohesion is one of the seven criteria of textuality (De Beaugrande and Dressler 1981), and ellipsis represents a cohesive link (Halliday and Hasan 1976), both in spoken and written text, and is one of the most common cohesive ties in English. Ellipsis is defined as the omission of linguistic elements that should be understood from the context, forcing the recipient to correctly fill in the grammatically permitted gap, for which in the case of substitution there is a replacement word, and in the case of lexical cohesion the same linguistic element or synonym will be used. Thus, people speak (write) elliptically without making much mental effort, and the recipients easily interpret and understand the elliptical sentences. According to Halliday and Hasan (1976: 142), the starting point in the definition of the term ellipsis is that "something is left unsaid", but not in the sense that the unsaid is not understood, or that it is unclear, on the contrary, the unsaid is understood. This means that the omitted element is missing from the surface of the sentence, but is present in its meaning.
When we consider this cohesive tie, we refer to linguistic slots in which information that is not given is assumed and listeners/readers understand the ellipted material based on the knowledge of their experience, or understand it from the context. This practically means that all spoken and written sentences are included in our understanding. Ellipsis is a reduction of a linguistic element, but we cannot define it simply as an ordinary omission of linguistic material. The stability of the text is established and maintained by the continuity of linguistic elements. The concept of continuity is based on the assumption (presupposition) of the mentioned linguistic elements. In this regard, the determination of ellipsis should emphasize the fact that this kind of cohesive connection is achieved only when there is a presupposition of what is left unsaid. Therefore, at the sentence level, ellipsis is best explained by relating it to the perspective of the functional sentence, especially the topic. A topic is a part of a sentence that carries information known from the previous text or context. This means that in elliptical sentences the topic can be freed from information that the sender thinks is known from the context or situation for the recipient of the text -although the interpretation of the sentence depends not only on the content of the sentence, but also on the recipient of the text -and thus there is no place for ambiguity. This we can see from the following example, where, similar to substitution, nominal ellipsis is a relationship within a text:

Example 11: "(…) Death is expensive, Miss Stella! And old Cousin Jessie's [] right after Margaret's [], hers! (…)" (SND, 26)
In most cases, the assumed element is present in the previous text. Thus, we can define ellipsis as an anaphoric relationship, and in this type of cohesive connection we return to the previous text for interpretation. The above example show that, where there is ellipsis, there is a presumption that something unsaid is understood. That is, we say that an element is elliptical if not all segments that contribute to its shaping are represented within its structure -all meanings are achieved with this structure.
In general, we can talk about ellipsis when there is something missing in order for the sentence to have a full and correct grammatical structure, so a sense of incompleteness appears in the structure, although it is not a question of semantic incompleteness. The main function of ellipsis is to achieve economy of linguistic expression by avoiding repetition. It contributes to clarity and emphasis; that is, through this cohesive tie it is possible to emphasize only what is new and important information.
By nominal ellipsis we mean the ellipsis of the noun phrase and the ellipsis within the noun phrase. The structure of a noun phrase contains a head and optional modifier(s): elements that can be found before the head in the phrase (premodifiers) or after the head (postmodifiers): Example 12: "You're the first person with his very own apartment who's ever asked me to marry him. I mean, I have been asked by other apartment owners but never to be a live-in wife. You're the first []. And you'll always be the first []." (SO,26) Example 13: "One second, the kids are tiny. The next [], they're old enough to go shopping by themselves. The next [], they're leaving home to start a montage of their own." (SO,51) From the examples, we see that the ellipted material can be replaced from the source of information, which is usually the nominal group that precedes the ellipsis. In the elided sentences from examples (12, 13), the nouns person/second are omitted; that we can easily reconstruct from the first non-elliptical sentence. First/next are numbers/determiners that normally have functions of premodifiers in non-elliptical structures. The function of the head is usually performed by nouns or pronouns and this function is always fulfilled. Modifiers can be determiners, numbers, quantifiers, adjectives and nouns. Postmodifiers are mostly prepositional phrases in form. Under the conditions governed by ellipsis, the head can be omitted, and then its function is taken over by some other modifying element from the phrase. This is exactly what constitutes a nominal ellipsis in English. Let us have a look at a few examples of nominal ellipsis as an illustration, before enumerating its subcategories: After deleting the head from the noun phrase, these determiners are promoted to its function in the ellipted phrase. The examples (14, 15) contain ellipsis of head words (point/hand), which can simply be identified from the clause that precedes it. Here we can notice that the elided structure has numeral for the head of the phrase after the deletion, to which the ellipted material from the clause that precede it can be replaced by simple addition.
Nominal ellipsis, in which the interpretation of the deleted material depends on what is in front of the elided structure in the sentence, are called anaphoric ellipsis. Also, there are structures where the interpretation of deleted material depends on what comes after the ellipsis; we call them cataphoric ellipsis. By their nature, elliptical noun phrases in English can be anaphoric or cataphoric, final or central, when it comes to their position in the sentence. Ellipted material can include only the head of the phrase or the head with its premodifiers or postmodifiers, if they exist. When deleting the head from the noun phrase, its function is taken over by those words that are otherwise premodifiers.
Within noun phrases in which ellipsis occurs, that is, where syntactic material is deleted, classification can be made based on the part of the phrase that has undergone deletion. According to frequency, the first case of deletion is the deletion of the head. Ellipsis of the head are the most numerous constructions among nominal ellipsis. In these constructions, where the ellipted material is the head, we noticed that after their deletion, the function of the head is taken over by premodifiers of the noun phrase. Analyzing the selected examples, we noticed that these premodifiers, according to the type of word, are in most cases cardinal numbers. In the elided structure, in relation to the non-elided one, the role of the head is assumed by: cardinal numbers, ordinal numbers, quantifiers, genitive forms, comparative or superlative forms, determiners, pronouns, adjectives. The largest number of excerpted examples of the head ellipsis has the cardinal number for the head of the elliptical noun phrase:  (SND,59) In the next group of examples (18,19,20), we listed those in which, after ellipsis, the head in the noun phrase becomes a quantifier. Among the quantifiers, much, many, and few are most often found: Other examples include determiners, pronouns and adjectives as heads of the elliptical noun phrases: Example 23: "There will be a nurse who comes one day and will not return the next []." (HWSG,42) Example 24: "I don't want to be friends with these women, anyway. (…) They are the old model and we are the new []." (HWSG,57) With the ellipsis of the sequence premodifier and head, we have the deletion of two elements within the noun phrase, the premodifier and the head, and the elliptical structure consists only of the determiner: Cataphoric nominal ellipsis, which are somewhat unusual in shape, are less frequently used than anaphoric ones. In example (26), we can notice that the elliptical noun phrases come at the end, and it is then possible to reconstruct the exact ellipted material from them. The pattern of deletion in cataphoric nominal ellipsis is absolutely identical to anaphoric ones, so we did not separate them individually: Example 26: "…and then Kirsten and I sleep in a room overlooking everything, the same view as the living room and porch above, with Beth next door and Toph sleeping -he sleeps like a dream; two[], three minutes and he's out -in a makeshift home for him we've made, with a curtain and a futon, out of the area between our bedrooms." (HWSG,63) The frequency of nominal ellipsis occurrence in English is high. Preference is always given to the elliptical over the non-elliptical structure whenever possible. The operation of deleting an identical structure is equally represented under the conditions of syntactic identity. There are rare examples that prefer repeating an identical structure, in both sentence constituents, rather than deleting it. Such examples are stylistically marked and have an emphatic role. Of all the nominal ellipsis, which we tried to present as clearly as possible in this chapter, the most common are the ellipsis of the head. Their number exceeds all other subclasses of nominal ellipsis. This data indicates that, if ellipsis occurs in a noun phrase, the entire phrase is always omitted rather than some of its parts: The text is a feature of modern linguistic studies. At the same time, cohesion and cohesive ties are one of the central issues. Cohesion analysis actually covers the semantic and grammatical connections in the text between sentences and parts of the text. Ellipsis as a cohesive tie plays one of the most important roles in the concept of texture and textuality.
In this paper, we presented ellipsis in a noun phrase, which, as all other types of ellipsis, leads to the economy of linguistic expression and, therefore, to textual cohesion and coherence. In that case, the recipient of the text can focus on more important parts of the text and new information. A full, non-elliptical answer, unless stylistically marked or used for emphasis, seems unnecessarily verbose, redundant, and unnatural. Analysis of the examples of nominal ellipsis indicates that this cohesive connection is very important for the formation of texts. Our research was mostly done on the basis of texts from dialogues, because unlike journalistic or administrative style, the use of this cohesive tie is much more frequent in the conversational and literary register:

Conclusion
In this paper, we have tried to offer an explanation about the cohesive tie of elliptical constructions that people use in language to communicate with each other fluently and without interference. We described the forms of cohesion, ellipsis, and the situational integration of these phenomena within the literary genre, and the ways in which their contents ensure textuality. Special attention is paid to the dominant form of these connections in literature, which are explained in detail through illustrative examples taken from such texts. We have concentrated on problems and issues related to reference and the general notion of cohesion, coherence and ellipsis. We have left several problems untouched for further research, and they are related to areas of semantics and syntax, such as issues of aspect, tense, modality, negation, adverbial modification, as well as relevant issues such as the influence of metaphor in the interpretation of discourse. Selected examples have shown that situational integration, and common knowledge of language and the non-linguistic world, are crucial for determining and interpreting that text where certain parts are missing: More recently, discourse analysis can be defined as belonging to interdisciplinary sciences and having a large number of points of contact with other disciplines, both in the areas of research and in its application. In this way, discourse analysis and text linguistics have become one of the most important areas of linguistic research since the seventies of the 20th century and are undeniably an important complement to systemic linguistics. We hope that this paper will also offer a modest but clear presentation of elliptical constructions as one of the cohesive means, which belong to the domain of general linguistic laws of linguistic economy. Above all, we hope that this paper will provide the reader with an insight into the functioning of the English language, but also encourage them to think anew about the nature of those complex cognitive and social phenomena that we call cohesion and ellipsis.