Abstract Schizophrenia is a mental illness that affects its patients' cognitive functions and different levels of language. Discourse is one of these levels that is severely affected by this disease. In this research, 11 schizophrenic patients and 11 healthy individuals were compared using the interpersonal metafunction in Halliday’s systemic functional grammar. To collect data, we used The Persian Aphasia Battery, Western Aphasia Battery, Boston Diagnostic Aphasia Examination, the story “Frog, where are you?” and questions such as “tell one of your good and bad memory, talk about Nowruz, etc.”. We employed SPSS software to analyze the data. The findings showed that there was a significant difference in the speech of patients with schizophrenia compared to healthy people in terms of the use of items such as polarity, modality, past and present tenses, declarative, WH-interrogatives and imperative speech functions, subject, complement and circumstantial adjuncts. However, no significant difference was found in the use of future tense, yes/no interrogatives and suggestion speech functions. According to the findings of this research and in line with previous studies, interpersonal speech systems can be used to provide a profile of the intensity of emotional and social disorders and incoherent speech in schizophrenic patients. Keywords: Schizophrenia, Discourse, Systemic Functional Grammar, Interpersonal Metafunction, Mood, Residue Introduction Language disorder refers to a mental malfunctioning in the understanding or using written, spoken, or other medium of language. A disorder as such manifests itself in the form of language (phonetics, morphology, and syntax), content of language (semantics), or the use and function of language in communication (pragmatics). These problems may be receptive, expressive, or a combination of both (Fogle, 2008: 249). One of the mental disorders in which language is affected as such and patients experience language disorders like the one mentioned above is schizophrenia. Schizophrenia can be defined as “a severe mental disorder characterized by a set of symptoms such as impaired thought and perception that has lasted at least 6 months.” (Arman et al., 2005, p. 86) The underlying cause of this disease has not been diagnosed completely yet and it has partly remained unknown. However, it is widely agreed that “schizophrenia is a heterogeneous disorder in which a set of biological, psychological, and environmental factors play a role in its occurrence” (Arman et al., 2005: 86). In 1980, Crow (1980) divided schizophrenia into two types of positive and negative schizophrenia (Andreasen, 1985: 380-382, as cited in Crow, 1980). Language impairment, thus, is one of the most obvious cognitive defects in schizophrenic patients (Kim et al., 2015:288) whose linguistic performance can be analyzed recruiting mainstream theories and approaches in today's linguistics, i.e., formal linguistics, functional linguistics, and cognitive linguistics. Systemic Functional Grammar is one of the mainstream theories of studying language pioneered by Michael Halliday (Rasakhmahand, 2013: 104). According to Halliday and Matthiessen (2014), language has different layers: 1) phonology and phonetics, 2) lexico-grammar, and 3) semantics. In Systemic Functional Grammar, clause is the main processing unit in lexico-grammar layer and represents different meanings. In this approach, Halliday talks about the context of situation and three components that create this context: field, tenor and mode. Each component of the context of situation has a representation in the semantics stratification, which is called metafunction. The three main metafunctions in Systemic Functional Grammar include ideational, interpersonal, and textual. As it concerns interpersonal metafunction, there are two components of mood and residue. Mood consists of subject, finite, and polarity. Residue includes the predicator, complement, and adjunct (Halliday & Matthiessen, 2014). According to Halliday and Matthiessen (2014), a communication with others always occurs based on certain social and personal experiences. These communicative experiences form the ground for interpersonal metafunction through which the clause is considered as an exchange. Given that language disorders are considered as one instance of schizophrenic symptoms, in this research, we compare schizophrenic patients’ discourse with healthy subjects’ discourse employing the interpersonal metafunction of Halliday’s Systemic Functional Grammar. The present study aims to provide justifiable responses to the following enquiry: What is the difference between the discourse of schizophrenic patients and the healthy control group in terms of the use of the functions of interpersonal metafunction? Materials and methods In this paper, we examined the speech of 11 schizophrenic patients (8 men and 3 women) and 11 healthy participants (8 men and 3 women). Schizophrenic participants were selected from Avicenna, Ostad Moharreri, and Hafez hospitals in Shiraz. The age range of participants in this group was between 24 and 60 years old (average = 40.18) and their average years of education was 11.27. These patients were selected based on the DSM-IV-TR diagnostic criteria for positive type schizophrenia and the psychiatrist’s confirmation. Participants in the healthy control group were selected from the adult community living in Shiraz. The age range of healthy subjects was between 21 and 60 years old (average = 40.27) and their average years of education was 11.54. To collect data, we used The Persian Aphasia Battery, Western Aphasia Battery, Boston Diagnostic Aphasia Examination, the story “Frog, where are you?” and questions such as “tell one of your good and bad memories, talk about Nowruz, etc.”. Interviews were conducted face-to-face between April 2022 and August 2022, with no time limit for answering. After transcribing the data, conversations were divided into clauses. Then, we determined all the functions of the interpersonal metafunction in the speech of each participant. These items included, polarity, modality, tense, speech functions, subject, complement, and adjuncts. Finally, the type, number, and percentage of each item was calculated and determined separately. SPSS software was used for data analysis. Discussion of results and conclusions The present study aimed to compare the functions of interpersonal metafunction in the discourse of schizophrenic patients and healthy control groups. Data were collected from 11 people with schizophrenia and 11 healthy people as a control group. We compared the two groups using Halliday’s Systemic Functional Grammar. The findings suggested that there was a significant difference in the speech of patients with schizophrenia compared to healthy people in terms of the use of polarity, modality, past and present tenses, declarative, WH-interrogative, and imperative speech functions, subject, complement, and circumstantial adjunct; but no significant difference was found between the two groups of subjects in the use of future tense and yes/no interrogative and suggestive speech functions. According to the polarity data, the percentage of negative clauses in the patients’ discourses (11.78%) was higher than in healthy people (9.66%). This, in line with the findings of Martins (2006; 2008) and Cohen (2011), can be attributed to the patients’ negative attitudes towards themselves, others, and the world. Therefore, this aspect increases the possibility of using negative words and negative clauses in their discourses. In line with Cohen (2011), the speakers’ frequent use of modality indicates uncertainty, lack of commitment, or confidence in the truth-value of the utterance; therefore, it poses a problem for the continuity of propositional information and fluency in patients. As a result, communication with others becomes difficult because the listener may fail to interpret the interpersonal situation and sequence of propositions. Patients also used a higher percentage of past tense than healthy subjects, because they linked the discussions to the past events. Furthermore, their views of others, events, and the world may have influenced the use of tenses in their conversations. Regarding the speech functions, a significant difference was found between the patients in declarative, WH-interrogative, and imperative speech functions. This was because the context and the type of discourse in which the subjects participated was conversation. Also, the patients used more WH-interrogative clauses than the healthy people, which indicates that they were looking for more information about themselves, their situations, and the subjects. The reason for this lies in the position that they hold for themselves. In addition, the patients realize themselves in the given situation and ask to the underlying reasons of what has come to them. The percentage of omitted subjects was higher in patients than in the control group. The more cases of omitted subjects and complements indicate that patients do not pay attention to the needs of the listener and the context of the conversation. Not paying attention to the needs of the listener is one of the reasons for producing a problematic and hard-to-grasp discourse. The higher number of circumstantial adjuncts and complements in the speech of patients compared to healthy people can also be attributed to the fact that the speech of patients is less frank in responding, and they are involved in expressing other events that automatically lead to longer speech. Another result that was found in this research was the higher number of incomplete or abandoned clauses in the speech of schizophrenic patients. It can be said that one of the reasons for speech defects in schizophrenic patients is these incomplete clauses, which go hand in hand with incomplete words/phrases, leading to different speech in schizophrenic patients. According to the data of this research, the Systemic Functional Grammar approach worked very well in examining the language of schizophrenic patients and healthy people and is a suitable approach for these kinds of research.