Referential Cohesion In The Construction Of Magazines English Language Essay

Texts are categorized according to different genres, which are placed within a context. For example, articles in magazines are arranged in a way of coherence so that readers are able to interpret from what is being read. Yet, there are argument in terms of the word ‘text’ and ‘discourse’. Like ‘cohesion’ and ‘coherence’ which needed to be distinguished, the inconsistencies of both terms have brought upon much confusion in studies done. Some researchers claimed to be doing text analysis but instead, have provided a study on discourse analysis albeit the basis of distinction of ‘text’ and ‘discourse’ as defined by Widdowson (1973; extracted from Widdowson, 2002); are the structure of sentences and usage of such sentences, respectively. But Widdowson (1978; Ibid.) himself did not maintain his definition and claimed later on, that ‘discourse’ consists of sentences with properties such as cohesion and coherence. Much confusion led to many researchers to conclude that ‘text’ is only used to refer to “physical arrangement of linguistic signals on paper” (Tadros, 1981), whereas ‘discourse’ is used to refer to every investigation on the structure (supra-sentential) of any range of spoken or written language. Otherwise, this distinction is deemed unnecessary.

Cohesion and coherence are analyzed based on individual texts and ergo, need to be distinguished. Cohesion correlates with the overt semantic relations whereas coherence deals with the relationship aspects of semantics, as well as pragmatics, within the text which are interpretable against the readers’ prior world knowledge (de Beaugrande & Dressier 1981). In other words, cohesion deals with the surface level of text, whilst coherence serves as the “underlying phenomenon” in the text. Coherence is otherwise known as texture, features the combination of semantic configurations which is made up of register and cohesion. Cohesion, being the main focus of Halliday and Hasan’s work, is said to be a display of existing ties, between a presupposed item and a presupposing item that occurred in the text; thus making the text cohere. That said, “Cohesion does not concern what a text means; it concerns how the text is constructed as a semantic edifice.”(1976:26). Cohesion, thus contributes to the “connectedness of the surface elements in the text” (Berzlanovich, 2008), and builds the structure of the text in form of coherent sentences; which in the case of this paper, cohesion will be analyzed in the context of a local fashion magazine’s article.

The way cohesion revolves around the text also depends on the varieties of discourse. Discourse here refers to “verbal communication in its situational and social context” (Ibid.). Since the roles of both cohesion and coherence occur interdependently with the type of genre, in some articles like academic discourse; it is found that conjunction is favored more as a cohesive link between sentences. Also, register-specific vocabulary in the text seems to be denoting as a primary factor for stronger cohesive effect, rather than general vocabulary (Teich & Fankhauser, 2005). As for narrative structured articles like those from the fashion magazines, previous studies have found that the dominant characteristic typically found in narrative texts is referential cohesion especially when participant chains, anaphoric links, temporal and spatial progression are being investigated (Fox 1987, Berzlanovich, 2008). Lexical items are necessary in the organization of a discourse in order to consider its meanings semantically and whether those sentences are in association with the intended meanings in the text. Thus, lexical items are categorized into two groups: ‘grammatical cohesion’ and ‘lexical cohesion’ (Table 1, Appendix 1). Grammatical cohesion encompasses reference, substitution, ellipsis and conjunctions, whilst lexical cohesion is classified into two types which are reiteration and collocation (Halliday & Hasan, 1976).

Previous studies have been published in abundance regarding textual cohesion and its devices. Nevertheless, so far to my knowledge, none of them has considered fashion magazine articles as a genre for analysis in the corpus of linguistic research, albeit many studies had been done in the area of language and gender but not from the linguistics aspect. Hence, an article from a local fashion magazine (Cleo) will be analyzed in relation to grammatical cohesion on one of its components- reference, which will be detected via the three dimensions as aforementioned: personal pronouns (determinative, possessive), demonstratives and comparatives (Appendix 2). Other dimensions of grammatical cohesion and lexical cohesion will be considered for future works, whilst in this paper referential cohesion is chosen to observe the linkage between sentences within the text, with the context in existence. The significance of this paper underpins the twofold objectives. Firstly, it is hoped that this study will be able to provide better understanding regarding the use of referential links in the way a cohesive text is constructed on the surface level of sentences. Secondly, the frequent use of cohesive links with functions and purposes, within sentences will help us gain insights in the aid of the readers’ interpretation of the text based on the type of genre examined (magazine article). As a result, the study on cohesion will help to put up with effective reading and writing altogether with word choices and paragraphing in texts and discourses. Finally, the present study provides a surface-leveled analysis of grammatical referential device; employed in the usage of English in this informal context.

Literature Review

Halliday and Hasan (1976) have provided the fundamental framework in the functional grammar studies; a lot especially on the studies of cohesion in many research conducted in this field. Although, other researchers like Werlich (1983) and Brinker (2005) have also made contributions in this field of study. Yet, none has offered detailed and distinguished differentiation in terms of cohesion and coherence. Halliday & Hasan (1976) only uses ‘cohesion’ whereas Werlich (1983) emphasizes in ‘coherence’ whereas Brinker (2005) points out that the distinction between cohesion and coherence is not necessary. On the other hand, researchers like de Beaugrande & Dressier (1981) offers explanation on the surface for both terms, and others followed suit. Most of their classification of cohesive devises branches out, primarily, into grammatical and lexical cohesion; in which the former will be the subject of concentration in this present study, primarily on ‘reference’.

Apart from that, Halliday also claims from the view of a functionalist in grammar; that language structure “is explained by derivation from” function and that he and Hasan has established a diversified functional/ text unity, mostly on structural elements like lexical repetition, lexical inclusion and many others which have contributed to the studies of cohesion.

According to Halliday and Hasan (1976), grammatical cohesion is divided into four different devices as follows (Querol 2004, Mohamed-Sayidina, 2010):

Reference: one element of the text is related to another one as aforementioned for its interpretation, with the use of pronouns (personal, comparative and demonstrative)

Semantic relation is required in this mechanism. It in the subsequent sentence refers to being normal.

E.g.: “Look at what being normal involves in this culture. A lot of it is what the rest of the world aspires to- …”

Substitution: the replacement of an item with another word to substitute a word or sentence. ‘Others’ here is used to substitute ‘some areas’ from being repeated.

E.g.: “Also, we have our own set of abilities and talents; in some areas, we’ll be above average, while in others…”

Ellipsis: the head noun, main verb, or even the whole clause is subsequently omitted in the following sentence. Consider this statement, whereby the making of sacrifices is omitted from the line that continues. Supposedly; “People who have extraordinary lives often have to make sacrifices that many of us wouldn’t consider worthwhile (of making the sacrifices)”.

E.g.: People who have extraordinary lives often have to make sacrifices that many of us wouldn’t consider worthwhile.

Conjunction: an invariable grammatical particle to connect two words, sentences, phrases and clauses to create discursive connections. There are several types of conjunctions namely coordinating, correlative and subordinating conjunctions.

E.g.: “…But when most of us wake up, life is much more prosaic, so we wonder what we’re doing wrong.”

Specifically, ‘reference’ as defined by Halliday, is a “participant or circumstantial element” which is being introduced within the text at one point; either as a reference point preceded for the following phenomenon, or functions as a basis for comparison. Referential cohesion can be realized in three ways: personal pronouns (determinative, possessive), demonstratives and comparatives (Appendix 2). But in the in-depth analysis of referential cohesion, it does not only matter that these items exist in the text. It also matters if they (personal pronoun, demonstrative or comparative) refer forward (anaphora) or backward (cataphora) to items within the text (endophora) or outside the text (exophora). Then again, readers might encounter some problems in relation to vague reference item, which neither refers to earlier or later items within the text nor to entities outside the text, even if confined within the context. Thus, self-referential occurs when readers interpret the text based on their cultural or world knowledge. This situation is also known as homophora. All these may function as clue items to provide more information to words and phrases (Muto, n.d.). But in the case of this paper, the main concern of the study is solely to detect the frequent use of referential cohesive links in order for the text to flow from one interpretation to the next; personal pronouns (determinative, possessive), demonstratives and comparatives are the focus analysis. Hence, the present study does not include in-depth analysis on cohesion elements like anaphora, cataphora, endophora and exophora, or even other elements like ellipsis, substitution and conjunction are omitted, because this current study deals solely on the surface level of referential cohesive links (reference) although it is suggested that future studies should looked into those aspects aforementioned. For future studies, M.A.K Halliday’s scope on references encompasses a wide range of aspects to be looked into especially in the field of systemic functional grammar where more elaborated, technical works of grammar classification could be done in relevance to this current study.

From lexical cohesion to grammatical cohesion, there are many research conducted in the field of cohesion and coherence, linguistically (Oliveria et.al, Klebanov & Shamir 2006) or pragmatically (Kruijff-Korbayova & Wolska 2008, Taboada 2004); yet, studies in this area are sensitive towards types of discourse, genre and text organizations (Berzlanovich, 2008). The study of this topic provides many helpful references and even contribute to the educational field (Muto, n.d., Mohamed-Sayidina, 2010), whereby analysis on lexical cohesion has been done on numerous students’ works. Also, the study on cohesion has led a group of researchers formulating a software called WordNet, to detect synsets (set of synonyms). As a result, more quantitative studies are conducted in the area of linguistics with the use of WordNet (Teich & Fankhauser 2004, 2005). Consequently encouraging mixed method amongst the research done in analysis or application of the device. In the analysis of cohesion, the study is presented either in systemic inventory of cohesion or instances of cohesion in texts (Kunz 2008). Then, variations also play a part in the determination of the purpose of study. There are four types of variation which are variation over time, across registers, across languages, and across originals and translations. All the analysis can be done via theoretical and example-based analysis, in-depth text analysis or empirical analysis. In the present study, the instances of cohesion are analyzed from the text in a particular variation of a register (fashion magazine); in regards to how often the distribution of referential cohesions occur, via the method of surfaced text analysis per se.

Methodology

Data Collection

An article was extracted from the local fashion magazine, namely Cleo. One article was adopted because it was due to pure intentions to merely detect the frequency of referential cohesive links throughout the text. In addition, these links are observed in terms of its occurrence in text apart from its connectivity within sentences. Fashion magazine was chosen because not many research have been done in this area of informal discourse, particularly in the feature of referential cohesion; most of the studies reflect on the academic (Mohamed-Sayidine 2010, Muto, n.d.), news (Oliveria et. al., n.d.), and formally written articles in magazines. No comparisons amongst genres were intended in this research; hence explains the reason for one article. Albeit comparisons amongst frequency of cohesive links used in various genres could lead to more future studies to be done.

The researcher analyzed the article to detect types of referential cohesion based on an adaptation of methodology from Querol (2004) and its frequency of occurrence for each of the component- personal pronouns (determinative, possessive), demonstratives and comparatives, in order to observe the patterns of sentence construction with these cohesive elements. Firstly, personal pronouns will be observed from the determinative and possessive forms respectively. Then, demonstratives will be divided into specific demonstrative and non-specific demonstrative. Lastly, comparatives will be in terms of general and particular comparatives. The results are listed in Table 2, 3, 4 and 5 respectively.

Results

Data Analysis

Table 2 presents the total of occurrences of 74 for personal pronouns which include determinatives and possessives; as well as 50 and 10 for demonstratives and comparatives, respectively.

Referential cohesion

Total occurrences (Percentages)

Personal Pronouns (determinative and possessive)

74 (55%)

Demonstratives

50 (37%)

Comparatives

10 (8%)

Table 2: Total of occurrences according to types of reference.

Then, in second place with total of 22 occurrences is calculated for determinative you/ yourself, preceded with determinative we /us, scoring the highest with a total of 24 occurrences. Determinatives such as they / them and it scored reasonably lower than the others with 9 and 10 occurrences respectively, whereas possessives pronouns like their/ theirs (3 occurrences) and its / our/ ours (6 occurrences) hardly occur throughout the text.

Personal Pronouns

Determinatives

(No. of occurrences)

Possessives

(No. of occurrences)

You/ yourself

22

She/ Her

–

He/ Him

–

They/ them

9

It

10

We /us

24

His

–

Her/hers

–

Their/ Theirs

3

Its / our/ ours

6

Table 3: Total of occurrences for Personal Pronouns (Determinative, Possessive).

As for Table 4, specific demonstratives such as this, these, here, those, that are found quite common (27 occurrences) within the text if compared to non-specific demonstratives like it and the (23 occurrences). Otherwise, the results for total occurrences of demonstratives’ are quite consistent throughout the text unlike personal pronouns’.

Demonstratives

Specific

(No. of occurrences)

Non-specific

(No. of occurrences)

This, these, here, those, that, there

27

It, the

23

Table 4: Total of occurrences for Demonstratives (Specific, Non-specific).

The same scenario with almost equal consistency of results is also applicable to comparatives; with 4 occurrences for general comparatives (otherwise, apart from, instead of, also) whilst particular comparatives scored slightly higher with a total of 6 occurrences for words like much more, as many of, better than, pretty great / good and it’s better to. Thus, from the findings it is prominent that the highest form of occurrences falls on the personal pronoun category with almost half of the text (55%) construction is conformed to this particular referential link.

Comparatives

General

(No. of occurrences)

Particular

(No. of occurrences)

Otherwise

Apart from

Instead of

Also

4

Much more

As many of

Better than

pretty great / good

It’s better to

6

Table 5: Total of occurrences for Comparatives (General, Particular).

Summary of Findings

Different types of references will yield various results depending on the purpose of construction within the sentences. Discussion in this section of paper will be based on analysis of the highest occurrences per se from findings of each table (Table 2, 3, 4, and 5). As can be seen in Table 2, the highest total of occurrences is scored by personal pronouns (determinative and possessive) (55%), followed by demonstratives (37%) and lastly, comparatives (8%). Personal pronouns are defined as the pro-form which are used to substitute the nouns. These pronouns are divided into determinatives (also determiner as the word modifier that determines the type of reference a noun/noun group has) and possessive (showing possession), albeit personal pronouns could sometimes function as a determinative but not in all cases. In the case of this study, the determinatives for personal pronouns are you/ yourself, she/her, he/him, they/them/ it, we/us. There has been much confusion which lies in differences of pronouns and determinative to the extent that some linguists viewed both of it to be closely related and that pronouns are in fact determiners without a noun / noun phrase. Nonetheless, a clear distinction between pronouns and determiners can be made with three main features that set the determiner apart from pronoun. A pronoun may be a determiner but a determiner cannot be a pronoun in the case of tagged questions.

Examples (extracted from the Cambridge Grammar of the English Language)

This is delicious, isn’t this?

Also, pronouns cannot appear anywhere else except in between the verb and particle for phrasal verbs whereas determiners could be placed after a particle (example A). In addition, pronouns are said to be in the form of “distinct genitive” but determiners are not (example B).

Example A:

pick it up (correct) pick up it (wrong)

pick this up OR pick up this

Example B:

This is mine/yours/theirs.

*This is all’s.

Personal pronouns reached the highest number of occurrences (74) as compared to demonstratives (50) and comparatives (10). The breakdown of the table into respective types of cohesion shows a higher reading generally in determinative (especially for we/us: 24 occurrences), instead of possessive. Such results are probably due to the type of genre and target audience which the text is focused upon. Hence the stylistic writing could be the choice of words used to manipulate readers’ mind. It shows solidarity and common grounds in terms of word use of we/us; which connotes the idea that readers and writers understand the situation and have common background knowledge for the context (de Beaugrande & Dressier 1981). Sometimes, the ‘we’ pronouns could be inclusive or exclusive if the writer wishes to include or exclude the audience from the text.

Extract #1:

“The pressure to be anything but normal starts out young when we’re told how important it is to get good grades, and by the time we’re in our twenties, we’re suddenly feeling that we should be better paid, better looking and better travelled than our peers-in fact, better equals best.”

Extract #2:

“A lot of our fear of normality comes from a misplaced notion that we should be excelling in all areas of our life, otherwise we’re failing. One issue is that often what we think are our goals aren’t actually our own dreams-they’re what we believe society wants us to achieve…”

Hymes (1967) once created the ‘Model of interaction of language and social setting’ and categorizes speech situation into eight components: namely setting, participants, form and content of text, intent and effect of ends, key, genre, medium, and interactional norms, which it is also stated that the text is formed also as a part of speech situation.

Later on, researchers (Halliday, McIntosh and Strevens) offered another explanation for derivation of features from a situation in a text. Three main headings are proposed: Field, Mode and Tenor which describes how a context is able in its function to determine the intended meanings from the text conveyed. The Field is the overall event which includes subject-matter as part of the element in order to function as a text in whole altogether with the ‘purposive activity’ of the audience. The Mode is “the function of the text in the event”, that means of its genre and types of discourse to be taken into consideration. Then, the Tenor is the type of “role interaction” in which the participants are involved in with the text whether it’s relevant or not. These features could help explain the situation in which the reader experiences with the text. Halliday (1976) also noted that the association of linguistic features with “configuration of situational features” as mentioned above (field, mode and tenor) forms a register in the text. Since the concept of cohesion as defined by Halliday (1976) is supported with register, the both can be effectively combined to constitute a text. Therefore, when the writer tries to form a purposive interaction with the reader based on the subject-matter of the text (Field) with relevance to its function and appropriateness (Mode), he or she tries to connect and convey the message to the public (Tenor) via the text read.

As for demonstratives in the present study of text analysis, they are employed into sentences when the meaning of the context is dependent on other things rather than the relative physical location of the readers at the instance. This is also known as the discourse deixis and in other terms “verbal pointing”. This, these, here, those, that are specific demonstratives whereas it and the are non-specific demonstratives. Demonstratives, in table 4, have shown an almost equal reading of frequency (27 for specific; 23 for non-specific). Although, it seems that specific demonstratives is used more often throughout the text if compared to its counterpart. Words like this, these, here, those, that are apparent throughout the text compared to it and the. Yet, there has been much confusion in terms of the usage ‘it’. Here, ‘it’ could be perceived as either a demonstrative or pronoun, depending on the semantic purpose of text. If ‘it’ is a pronoun, it refers to the object of unknown gender or neuter. But if ‘it’ functions as a demonstrative, the meaning of the word needs to fall back on the logic behind the text. For example, “But, somehow, it’s become a byword for mundane and pedestrian.” Of course, ‘It’ in this sentence does not refer to a human being but the situation mentioned in the preceding sentence. The reader cannot simply comply with the meaning of word without referring to the context. The ‘it’ substitutes a noun and this replaced noun is also known as the pronoun’s antecedent. In cases of anaphora, ‘it’ becomes indecipherable without its context and thus the process of determining the intended meaning of antecedent. In addition, ‘it’ might also be dummy pronouns, which mean that none of the semantic relations are needed in relation to the context. For instance, “It’s an understandable pressure to feel under…”

Extract #3:

“It’s the bright, shiny moments that we feel like broadcasting, so it’s easy to get the impression that someone’s life is fabulously fast-paced and exciting, but that’s because it is only interesting to document that thrilling fraction.”

The interpretation of the text relies on the reference items within sentences, which can be interpreted in two ways; either the item is identified with referent “in question” or it is being implicitly compared to a referent. When identification is involved with the interpretation of the text, the reference item must always be specific and deitic (Halliday & Hasan, 1976) because the identifying function for cohesive purposes must also be specific. Hence, this, these, here, those, that are more commonly used throughout the text. These demonstratives could function as modifier and head of the sentence. Based on Extract #4, there in the sentence, “There’s some kind of Hollywood ideal out there…” functions as the Head of the former word, whereas the latter there works as a modifier for out. Apart from that, these words could apply anaphorically to mean the previous things which have been said or mentioned before between the interactions of the speaker/ writer and interlocutor/reader (Extract #5).

Extract #4:

“Dr. Chris Day, psychologist and author of 52 Secrets of Psychology (Dennis Jones & Associates), says that many young women feel that a normal existence isn’t good enough. “The pressure to be living this fun ideal life, knowing amazing people, doing amazing things. There’s some kind of Hollywood ideal out there that women aspire to. But when most of us wake up, life is much more prosaic, so we wonder what we’re doing wrong.”

Extract #5:

“Look at what being normal involves in this culture. A lot of it is what the rest of the world aspires to-sun, lifestyle, healthy food, friends, lots of public holidays, travel. Normal here is pretty good.”

Finally, comparative comes in the form of adjective or adverb, mostly to denote the property and degree of quality in which it is either greater or less in comparison of both entities. It is used in the context of subordinating conjunction (e.g. than, as..etc). Also, in comparatives, particular comparative occurred more often (6) compared to general ones (4). Particular comparatives refer to comparison in respect to quantity or quality. It is also expressed with aids from adjectives or adverbs. Examples of particular comparatives throughout the text are much more, as many of, better than, pretty great / good and it’s better to.

General comparatives include both similarity and differences in comparatives, in terms of likeness and unlikeness whereby comparison of such is irrespective of any particular property. Such general comparatives take forms in phrases like otherwise, apart from, instead of and also. Since comparatives are used to make comparison of the degree in both entities, it is quite common in narrative and expository writings. Regarding the text chosen for this study, albeit its structure seems more alike an expository; but there is a party (psychologist Dr Chris Day) whom narrates and advises on the pertinent matter. Thus the text is written in a narrative style. Using more comparatives to make differentiation of pros and cons, phrases like apart from, instead of, much more, etc is apparent throughout the text.

Limitations

Much research can be done on the analysis of variations across languages, registers and even original articles and translation ones. In this current study, only general text analysis is conducted whereby the author decides to omit the theoretical and empirical analysis due to time constraints and lacked of sufficient resources. The author is also restricted in terms of systemic functional grammar knowledge; thus planned to look at a more generalized grammatical cohesion as her basis for the study. For suggestions, a more elaborated and in-depth text analysis on systemic functional grammar perspectives could be conducted in another study for reference in the future. Also, the analysis would not be precise because only the author herself is analyzing the text. Probably, another person could help in second analysis of the same text to make comparisons between both findings (number of occurrences). Also, no software is used to detect the readings; neither SPSS nor WordNet is applicable to this study due to the nature of this study, partly due to the major time constraints that the author decides on a simpler form of method for analysis. Apart from that, more articles are needed for a substantial analysis to justify the substantive issues on grammatical cohesion.

Perhaps, more than one genre (magazines, newspapers, etc) and different aspects of cohesion could be looked into; including the elements of anaphora, endophora, exophora and others to name a few. Not only that, such article from an informal context of pure reading pleasure could result in a more informal stylistic writing. Moreover, the quality control of the article also depends on the publishing company; which is vital in this case of the study. An error has been detected during analysis- note that in paragraph 10 of the text- the sentence should read as ” happier than those jetsetting…” instead of “happier that those jetsetting”. This could affect the reading of frequency if error not detected, reflecting the unreliability of the text.

Conclusion

Whilst the present study is focused merely on the grammatical aspect of referential cohesion, other dimensions of this study could be considered in terms of variations, analysis and types of discourse. In addition, the focus on reference in this study could also be expanded for future studies since the scope is widely ranged in this topic of linguistics. Methods could also be diversified with the help of software such as WordNet and even SPSS for quantitative approach. Textual patterns, linkage, ties and chains could be observed from various perspectives either from a formal or informal context.

Fashion magazine was chosen because not many research have been done in an informal context of reading pleasure; most of the studies reflect on academic, news, report and formally written articles. Apart from ‘cohesion’, ‘coherence’ could also be distinguished from the former with more studies done in various discourses, not to mention on the aspect of lexical cohesion as well since both ‘cohesion’ and ‘coherence’ are important characteristics in prominent texts (Celce-Marcia & Olshtain 2000). Such study on the way cohesion links from one sentence to the next might provide suggestions in the comprehension of effective reading and writing, apart from word choices at semantic level in stylistic writing. Nevertheless, results found in the tables as aforementioned have shown reasonable occurrences across the genre to better understand the cohesiveness of text with help from referential links, whilst the frequency of these links occurred within the text indicates that this particular type of genre-narratives-involves particularly a significant use of pronouns in the making of the article. Ultimately, the understanding of cohesion provides a more comprehensive approach towards discourse analysis and text reading.

Appendix 1

Grammatical

Cohesion

Reference

Implies that the information is to be retrieved through the reference item is the referential meaning (pronouns and determiners – personal, demonstrative and demonstrative)

Substitution

Refers to the process of replacing one expression by another (e.g. Did Susie finish her homework in time? – I don’t think so.

Ellipsis

Designates the process of substituting something by nothing (e.g. How many potatoes do you want, Sir? – Four [], please.

Conjunction

Covers the use of adjunct-like elements to mark the semantic relationships (e.g. He didn’t study much. Nevertheless, he passed all his exams.

Lexical

Cohesion

Reiteration

General*

Repetition

Suggests the lexical item is repeated

Synonymy

Relates lexical items which have the same meaning

Antonymy

Constrasts between a term which is the opposite of another

Hyponymy

Relates specific and general lexical items, such that the former is included in the latter

Meronymy

Relates parts and wholes

Table 1: Categories according to Halliday and Hasan study (1976) (adopted from Oliveira et.al)

*Another device for lexical cohesion called collocation, which involves equivalence, naming and semblance, has been systematized in Hasan’s later study (1984) to compensate for the loss of clarity, yet it has not been developed to the same extent as reiteration.

Appendix 2

(Extracted and scanned from Halliday & Hasan (1976); Cohesion in English.)

Personal reference

personal.jpg

Demonstrative reference

demonstratives.jpg

Comparative reference

comp.jpg