Critique on Halliday’s Systemic View in Systemic-Functional Linguistics
FAN Fan
I.	Introduction
Systemic view is an innate essence of Halliday’s linguistic theory. As the "father of modern linguistics", Saussure has already put forward the concept of language system. The linguistic system has taken on different meanings in Copenhagen, London and Generative Linguistics. Halliday points out that the concept of system (from which systemic theory gets its name, Halliday, 2003, p. 186) he uses is derived from his teacher J. R. Firth. Early in the first phase of Halliday's theory he comes up with the concept of systems, and the importance of the system came to light in the middle of Halliday’s theory. He begins to bring systematic research to the forefront in the mid-1960s. Some scholars bear certain doubts toward Systemic-Functional Linguistic theory based on his systemic views. This essay tries to state Halliday’s Systemic view revealed in his theory and some of its merits and flaws. Given critiques from this perspective are basically rare and systemic view is so important in SFL, this paper hopes to explore in this field as a preliminary and superficial attempt to arouse more critiques and researches from higher angle to provide more solid macro theoretical guidance for the future development of systemic functional linguistics.
II.	Difference Between ‘Systemic’ and ‘Systematic’ Approach
Halliday’s view of language as a system is a systemic not systematic system. As Halliday has stated, ‘the name ‘systemic’ is not the same thing as ‘systematic’. (Halliday, 2003, p. 180) Systemic means ‘affecting or connected with the whole of something, especially the human body’ (Oxford Dictionary), ‘of or relating to a system, especially as opposed to a particular part’ (Webster Dictionary), ‘something that is spread throughout, system-wide, affecting a group or system, such as a body, economy, market or society as a whole’ (The New Oxford Dictionary). It is different from systematic which means ‘done according to a system or plan, in a thorough, efficient or determined way’ (Oxford Dictionary), ‘done or acting according to a fixed plan or system, methodical’ (Collins Dictionary), ‘a. relating to or consisting of a system, presented or formulated as a coherent body of ideas or principles, b. methodical in procedure or plan, c. marked by thoroughness and regularity, d. of, relating to, or concerned with classification specifically taxonomic’ (The New Oxford Dictionary). 
Hence the word ‘systemic’ emphasizes on the relationship with the whole. For example, if an organ of a person has systemic ill effect it means not only the part of the organ itself has something wrong, but the whole body might be unworkable. Or the phrase ‘systemic infection’ means the whole body is infected, like the disease ‘systemic lupus erythematosus’. Or the ‘systemic circulation’ means circulation of the whole entity. However the word ‘systematic’ emphasizes the plan or efficient procedure in relation with methodology and classification. For example ‘science is systematic’, ‘a systematic approach to solving the problem’, ‘the search is carried out systematically’, or ‘they do their business in a systematic way’ which all involve do things in a planned way following certain procedures. 
Therefore ‘systemic study/approach’ and ‘systematic study/approach’ mean quite different things. The former views studying object as an inseparable and continuum system consisting of many subsystems among which are many commonalities and all the whole systems is connected in inseparable continuum, while the latter analyzes the whole system structurally part by part. Halliday’s Systemic-Functional Linguistics adopts language system as comprehensive which is ‘concerned with language in its entirety’, therefore whatever is ‘about one aspect is to be understood always with reference to the total picture. At the same time, what is about any one aspect also contributes to the total picture’. Because languages are not designed but evolve, and ‘evolved systems cannot be explained simply as the sum of their parts’. Halliday stands that ‘traditional compositional thinking about language needs to be… complemented by a ‘systems’ thinking… to seek to understand the nature and the dynamic of a semiotic system as a whole’ (Halliday, 2014, p. 20).
III.	Systemic View in the Development of Halliday’s SFL
Eirian C. Davies has generalized the development of Michael Halliday’s work in SFL over more than sixty years and points out the six stages in Halliday’s thought. The development of the model includes: ‘beginnings of the model; Systemic-functional linguistics: the lexicogrammar; SFL and social semiotics; return to the lexicogrammar, grammatical metaphors; expansion of systems; and return to Chinese linguistics’; among which the former four are the main stages and all of which contain considerable overlapping in between. The first, second, third, and fourth stages raise and expand the idea of system. 
The first stage marked by Halliday’s article ‘Word’ in 1961 is mainly to put forward and study the grammar of ‘Scale and Category’. Halliday believes that language can be divided into ‘substance’, ‘form’ and ‘situation’. ‘Categories’ are used to express the different classification of language components in different language planes and the relationships among these categories. There are four categories for language: ‘unit’, ‘structure’, ‘class’ and ‘system’. ‘Scale’ is used to describe the relationships among categories, including ‘rank’, ‘delicacy’, and ‘exponence’ which is later termed ‘realization’. Among the categories, ‘system’ is the most important category that represents a combination of meanings available on a unit component. The components are called ‘terms’. The language user must make a choice in the system according to the meaning one wants to express. When the speaker has made one choice, which means he/she has excluded other choices. The speaker selects the meaning within the system of different components to form a linear structure, which, together with the paradigmatic relation embodied in the choice, combines to form a discourse expressing a specific meaning.
Since the late 1960s, Halliday's language theory has entered the second stage which sees a strong emphasis on system, at the expense of structure to some extent. ‘Theme’ and ‘transitivity’ are chiefly expressed in terms of system networks in Halliday’s three ‘Journal of Linguistics’ articles in 1967 and 1968. It was from this stage that his theory of language came to be known as systemic-functional grammar. Before the second stage Halliday used to think of language as consisting of three planes, but now the term ‘plane’ is replaced by ‘stratum’. The system in this stage includes not only the various meanings described in the term, but also all the combinations of meanings in the language. In Halliday's opinion, language is a sign system composed of various subsystems, but the so-called sign system should not be regarded as a system of signs, but as a systemic resource for meaning. His concept of grammar is broad grammar, including semantics, lexicogrammmar and phonological cohesion system. The language system consists of three levels: semantics, lexicogrammmar, and phonology. There is an embodied relationship or the stratification among these three levels: semantics (meaning system) is embodied by lexicogrammar (wording system), and lexicogrammar is realized by phonology (sound system).
The focus of system doesn’t show much in the third and fourth stage of Halliday’s theory except ‘a computational approach to the grammar using systems’ and ‘the concept of ‘realization’ and different levels (strata) of abstraction, which represents a development of the scale of exponence’ (Davies, p. 4) in Halliday’s work in the third stage. In this stage Halliday broadens his study to social semiotics and publishes his complex ideas to do with it. Relationship between language and social class and child language development are illustrated. In the fourth stage Halliday expresses largely in terms of structure in the first two editions of Introduction to Functional Grammar in 1985, and 1994. 
In the fifth stage of development Halliday focuses on the reinterpretation of the concept of system, which represents a further development of stage two’ approach. The system has been placed in a more important position. The publication of IFG in 2004 by the third edition in collaboration with Christian Matthiessen expands the idea of system. There is much more explicit account in terms of systems in this version than in the former two earlier editions. And the last stage Halliday returns to the study of the Chinese language. 
IV.	Systemic Approach in Halliday’s SFG
According to Halliday, ‘a system is a set of options with an entry condition: … a set of things which one must be chosen, together with a statement of the conditions under which the choice is available’ (Halliday, 2003, p. 180). And Halliday’s ‘formulation is in terms a ‘systemic’ description in which the grammar takes the form of a series of ‘system networks’ (Halliday, 1966, p. 37). ‘The system network is the grammar’. And ‘the grammar is based on the notion of choice’ (Halliday, 2003, p. 180).
The grammar of any language can be represented as a very large network of systems, and arrangement of options in simultaneous and hierarchical relationship. The network is open-ended. To the left of each arrow is the entry condition; to the right, joined by a brace, is the set of two or more options which make up the choice.  
(Halliday, 2003, p. 180)
A system is a set of features one, and only one, of which must be selected if the entry condition to that system is satisfied; any selection of features formed from a given system network constitutes the 'systemic description' of a class of items. Such a 'selection expression' is then realized as a structure, the structural representation being fully derived from the systemic; each element of the structure is a point of entry into a further system network, so that constituency is based on the concept of 'rank', with minimal bracketing. 
(Halliday, 1966, p. 37)
The theme of ‘The Ninth Systemic Workshop’ is ‘Applications of Systemic Theory’. After the workshop Halliday enumerates what systemic theory is about:
• interpreting the nature, the functions and the development of language
• understanding the role of language in expressing, maintaining and transmitting the social system, in home, neighborhood, school and other domains of the context of culture
• helping people to learn language, whether mother tongue or other languages, whether children or adults; and to use language effectively as a means of learning and in a variety of contexts of situation
• helping people to overcome language disorders, educational or clinical: 'slow learners', 'backward readers', aphasics, the mentally handicapped, etc.
• understanding the nature of discourse, and of functional variation in language (register); studying particular types of discourse (classroom, medical, etc.) for practical purposes such as the training of teachers and of specialists in the field
• understanding the nature of 'value' in a text, and the concepts of verbal art, rhetoric, and literary genres; gaining access to literature through the careful study of such texts
• using computers to analyse and generate discourse; developing a grammar for decoding and encoding, and a semantic representation to direct and interpret the grammar
• exploring a range of practical activities where language is involved: forensic issues, readability and complexity measures, communication in institutional settings and so on
• relating language to other semiotic systems and to the ideological patterns of the culture
This is a rather broad spectrum of doings, and it reflects the broad foundation on which systemic theory is built. 
(Halliday, 2003, p. 185-186)
In Halliday’s theory, he points out that what a text mean can only be explained by ‘relating it to the linguistic system as a whole’ (Halliday, 2014, p. 3). The system of language contains subsystems of phonological, graphological and lexicogrammatical constituencies with patterns distributed across units of varying size, ranging from the largest units to the smallest units. In the constituency of the hierarchical rank scale the smaller units combine together to form a higher rank. In the sub-system of sounding (phonology), the largest units carry melodic patterns and the smallest carry articulatory patterns. In phonological constituency, sequences of phonemes combine to form a syllable, syllables to a foot (rhythm group), and feet to a melodic line (tone group). Articulatory features are associates with phonemes (vowels and consonants) - smaller phonologic units that can combine to form syllable. And syllable is the gateway that it realizes prosodic features of intonation and rhythm associates with larger phonologic units. Articulation is arbitrary while prosody is related to meaning. The graphological rank scale of modern English writing includes four ranks: sentence, sub-sentence, word and letter. ‘The sound system and the writing system are the two modes of expression by which the lexicogrammar of a language is realized’ (Halliday, 2014, p. 7) The lexicogrammatical constituency consists of clause complex, clause, phrase or group, and word. Morphemes and word morphology is not being dealt with systematically in Halliday’s book because he doesn’t suppose composition should be allowed to dominate the thinking about grammar. 
‘The clause is the central processing unit in the lexicogrammar’ (Halliday, 2014, p.10). The principal network of systems of the clause: theme, mood and transitivity are elaborated in Halliday’s theory in accordance to the three metafunctions of language: textual, interpersonal and ideational [logical~experiential] metafunctions. A clause as a unit combines three lines of meanings – clause as a message, as an exchange, and as a representation in reference to the three distinct kinds of metafunctions. ‘For example given the entry condition clause there are three simultaneous choices to be made: from theme, mood and transitivity (For speech a fourth choice is added: information.)’ (Halliday, 2003, p. 180). 
‘Theme functions in the structure of the clause as a message’. (Halliday, 2014, p. 83) The significance of any functional label lies in its relationship to the other functions with which it is structurally associated. It is the structure as a whole, the total configuration of functions that construes, or realizes, the meaning. The theme systems concern the message is given or new and elaborated to be mark or unmarked textual, interpersonal and topical themes. The mood systems concern the types of mood, whether subject or finite, and residues, whether the residue contains predicator with complement or adjunct. ‘The transitivity systems are concerned with the type of process expressed in the clause, with the participants in this process, animate and inanimate, and with various attributes and circumstances of the process and the participants’. (Halliday, 1967, p. 38) The types of process systems include material process (processes of doing), mental process (processes of sensing), relational process (processes of being), verbal process (processes of saying), behavioral process (processes of behaving) and existential process (processes of existing or happening).
V.	Merits and Imperfections
Systemic-functional linguistics provides an important theoretical basis for language researchers. At the same time, like all other theories, it has both advantages and some shortcomings. Based on its systemic view, some scholars doubt the adequacy to description in Halliday’s SFL. They don’t think Halliday’s system is adequate enough to describe all languages and argue whether his description can fully capture all language features. Chomsky is among the scholars with the doubts. And within the language system since it is systemic but not systematic, the interfaces between different sub-systems are unclear, for instance the interface between the larger subsystems such as between grammar and meaning and the clear distinction between elements within small subsystems such as within the theme system the distinction between theme and rheme. 
Chinese scholar Yongsheng Zhu and Shiqing Yan also point out some defects in their Reflections on Systemic-Functional Linguistics (2001). Firstly they think, in terms of the systemic description, given our natural language system is too complex, it is not easy to describe any language comprehensively by using Halliday's description rules. For instance Systemic-Functional linguistics has made a great deal of description of English subsystems, but due to the complexity of the language itself, it is hard to say that this description is exhausted. And Halliday’s definition of the subsystem networks depends much on a number of variables. However the rules of deciding when to use the variables and which variables should be considered are not well given. This makes the range of subsystems can be smaller or larger. On the one hand, it gives language researchers great freedom in description; on the other hand, it is hard for them to come to an agreement on the description of the same subsystem in the same language. Secondly, the description of language system in SFL so far has not fully reflected the actual application of language. It is an arduous task to describe the dynamic mode of language system objectively and thoroughly. Halliday's linguistic theory is roughly equivalent to Chomsky's ‘external language’ or the application of language. Its purpose is to study the role and manifestation of language in communication and fully reflect the actual operational process of language on the basis of the description of language system. But even if all the subsystems of language are described objectively and adequately, it seems this description is not helpful for the understanding of the continuous operational rules or laws of language. Thirdly, the study of language systems in systemic-functional linguistics is still far from the ideal goal that the research results must be applied by other disciplines, especially by empirical applied science, which is a starting point of Halliday's language research. Halliday attaches great importance to the application of his language theory in artificial intelligence and natural language processing. Systemic-functional linguists are fully aware of the complexity of language use and the importance of contextual and cultural factors. Their results of the proposed theory are closer to linguistic facts than those of other schools of linguistics that ignore such factors. However despite Halliday’s focus on the influence of context and other factors on the process of language communication, it is far from enough to be directly used in the man-machine dialogue processing field. Last but not least, some scholars doubt where the boundary of Halliday’s system is.
However, a study needs to be fully understood by its theoretical perspectives, starting point and nature or properties. The primary purpose of SFL is to scientifically describe language. Things can’t be undone just because it is hard to do. Hence Halliday’s systemic description can’t be denied with its great contribution just because language is hard to describe. And the hardness is not caused by the complexity of the way to describe language, but the very nature of language itself.
Further, the argument that the distinctions among networks of subsystems and within subsystems are unclear shows some scholars misunderstanding toward the idea of fuzziness and they got too used to the traditional systematic way to view or analyze system. The systematic way of analysis usually presumes a system to be static with clear boundary and manages to divide it to different parts in accordance with certain ‘clear’ rules. Thus it can do the work as so-called understanding something thoroughly to the very ends. In this way the system is abstracted from what in the real world to be endowed with boundaries. However according to the theory of system a system is with open ends and can be dynamic. And abstraction is not the only way in the scientific analysis, especially of language which is complex and dynamic in social use with multiple perspectives. If related with society, the boundary is blurred and in that circumstance the study of language can’t only follow the clear cut dividing rules or binary rules. And fuzziness doesn’t mean unclear. It’s just not easy to decide as the clear cut rules and it does not work in that way. It’s another way to view language and as matter of fact closer to the nature of language. And it shouldn’t be a severe problem if it needs researchers’ work to define in language analysis given up until now there is no linguistic science that doesn’t needs researchers’ endeavor.
Chomsky’s Universal Grammar aims to analyze I-Language. It is the study of innate language device and internalized attributes of all human beings. He abstracts language as pure without real social factors’ influence. In comparison Halliday’s study is closely related with social functions, the E-Language. If developed to their perfectness, Chomsky’s theory would be like to be certain what is the innate device and what are the exact factors that cause human beings’ to generate languages universally just like scientists are sure about the fixed number of the pairs of human beings’ genetic chromosomes. And Halliday’s theory would be certain to know how exactly in the society each people look like outwardly. From the ironic extreme cases it can be deducted that knowing all the people’s appearance is not a necessary scientific result, but illustrating certain ways how genes would influence human beings appearance are. Therefore, the description of Halliday’s systemic theory on language is good and scientific enough while including every ‘end unit’ of language is not necessary or scientific. Hence it is unrealistic to expect it to be as definitive as UG or to include all language units or phenomena in its description.
Halliday has emphasizes ‘the unit of lexis and grammar, as the two poles of a single cline, or continuum’ (Halliday, 2014, p.64) which shows the concept of ‘continuum’ is important in his systemic view. Maybe it arises from Chinese philosophy of ‘Tai Ji’ that interfaces between layers are not clear cuts, which is different from most other western scholars to view things with ‘walls’ between things. It is due to the great connecting and inclusive strength of his language system that it is not limited within its own boundary and is applicable to many fields such as the application in children’s language development, language and society, register, language teaching, discourse analysis, stylistics, translation, and even beyond normal adults’ language to other semiotic systems, etc. that other schools of linguistics can’t attain. 
Nevertheless, as a science of language, Halliday should make his theory more profound not only to answer the ‘what’ question: what is the description of language; but to be more sufficient in answering the ‘how’ question: how to measure to describe language clearly; and especially focus further on solving the ‘why’ question: why language is described in this way. Otherwise the theory would be trapped in endless descriptions of language. The system seems to lie in descriptive categorization however lacks logical attribution. For example, keys to questions like ‘why the metafunctions of language are the three’ and ‘why certain variables are chosen but not others’ need to be offered. Furthermore, its language system should involve the deeper implicative meaning of language, and the social, psychological and cognitive intentions behind language. Therefore, how to integrate these factors into the study of Systemic-Functional Linguistics and make efforts to be applied to the future new field of artificial intelligence needs to be explored.
VI.	Conclusion
In view of the above illustrations of Halliday’s systemic view in SFL and the current critical reflections, this paper tries to fully affirm Halliday's systemic thought, at the same time, points out its imperfections, so as to obtain a comprehensive understanding of it. There is no doubt of Halliday and other scholars’ towering and continuing achievements and systemic linguistics is bound to be applied more widely.

References
Changling Zhao and Wei He, ‘A review of Lise Fontaine, Tom Bartlett and Gerard O’Grady (eds.), Systemic functional linguistics exploring choice. Cambridge: Cambridge University Press, 2013. 533pp’, Functional Linguistic, 4 (2017), 1-4.
Eirian C. Davies, ‘A retrospective view of Systemic Functional Linguistics, with notes from a parallel perspective’, Functional Linguistics, (4) 2014, 1-11.
Fuzhen Si, ‘Chomskyan Linguistics and the Scientific Methodology in Linguistic Study’, Asian Social Science, 5 (2009), 82-86.
J. R. Martin and Michele Zappavigna, ‘Embodied meaning: a systemic functional perspective on paralanguage’, Functional Linguistic, 6 (2019), 1-33.
Lulu Cao, ‘On systemic-functional linguistics’, Journal of Educational Institute of Jilin Province, 4 (2010), 64-66.
M. A. K. Halliday, Halliday’s Introduction to Functional Grammar (4th Edition) (London: Routledge, 2014).
M. A. K. Halliday, ‘Notes on Transitivity and Theme in English: Part 1’, Journal of Linguistics, 3 (1967), 37-81.
M. A. K. Halliday, On Language and Linguistic, ed. by Jonathan Webster (London: Continuum, 2003).