A Critique on Universal Grammar Li Na Content Abstract Ⅰ. Introduction Ⅱ. Research Background ⅰ. Universal Grammar ⅱ. Universal Grammar and Particular Grammar Ⅲ. Reflections on the application of UG on SLA Ⅳ. Confusions in UG ⅰ. Methodology ⅱ. UG and Rationalism ⅲ. Logical Problems ⅳ. Limitations of UG Ⅴ. Conclusion References Abstract: Chomsky proposes that languages are ruled by a set of innate universal grammars. The paper attempts to present what Chomsky means by UG. Second part devotes to the application of UG as well as reflections on the accessibility of UG to SLA. Controversies and confusions in UG in terms of research methodology, the relationship between UG and Rationalism, confusing logical problems and limitations of UG will be covered in part three. The conclusion part highlights the values of UG in spite of its limitations. Keywords: Universal Grammar; Second Language Acquisition; Limitations Ⅰ. Introduction “In the late 1950s, Noam Chomsky tried to find a new way to explain language when he found the limitations of the classification of structural elements of language for its distribution and arrangement”(Hu, 2006). From this practice, Chomsky gradually set up his famous Transformational-Generative (TG) grammar. His publishing Syntactic Structure in 1957 marked the beginning of the Chomskyan Revolution which led structuralism to a new era. Theoretical linguistics has undergone a great and influential change, with the focus of linguistic studying shifting from the description of language to the investigation of language nature. By a generative grammar, Chomsky simply means “a system of rules that in some explicit and well-defined way assigns structural descriptions to sentences”. He believes that “every speaker of a language has mastered and internalized a generative grammar that expresses his knowledge of his language.” “Thus, a generative grammar attempts to specify what the speaker actually knows, not what he may report about his knowledge” (Chomsky, 1965). Since its birth, TG Grammar has seen five stages of development. The Classical Theory proposed in Syntactic Structure challenged the popular American Structuralism regarding such linguistic questions as a cognitive basis, research object and research methodology. The Standard Theory (ST) focuses on how semantics should be studied in a linguistics theory. According to ST, generative grammar consists of three components: syntactic, phonological and semantic, of which syntactic component is the core and phonological and semantic component explanatory. In the Extended Standard Theory (EST), Chomsky completely shifts the semantic interpretation from the deep structure to the surface structure. The Revised Extended Standard Theory (or Government and Binding) develops the discussions in EST and supplements a principle system on the basis of a rule system. The fifth period, the Minimalist Program (MP) attempts to revise the previous theory. TG grammar is neither a prescriptive grammar, which attempts to lay down rules for “correct or standardized language, nor a descriptive grammar, which tries to describe and analyze the language people actually use, but it intends to investigate something deeper—the foundational principles that make language possible across all of humanity. The core idea of TG grammar is that all human beings are born with an innate capacity for language—and that this capacity shapes the rules for what is considered “correct” grammar in a language. Of course, any theory in an academic setting will have challenges, comments, and criticisms. The idea of the innate language capacity—or a “Universal Grammar”—is not accepted by all linguists. Some believe, to the contrary, that language is learned and therefore based on certain constraints. The paper attempts to present what Chomsky means by UG. Second part devotes to the application of UG as well as reflections on the accessibility of UG to SLA. Controversies and confusions in UG in terms of research methodology, the relationship between UG and Rationalism, some confusing logical problems and limitations of UG will be covered in part three. The conclusion part highlights the values of UG in spite of its limitations. Ⅱ. Research Background ⅰ. Universal Grammar Based on his observation that children can learn their native language very fast and with little effort in very different environments, Chomsky believes that language, as the basic intellectual component of human being, is somewhat innate, and children are born with what he calls a Language Acquisition Device (LAD). Why human beings can acquire the language is attributed to their special brain materials and structure. When children are born, they have been endowed with linguistic knowledge in the delicate human brain. Those pieces of linguistic knowledge are general and abstract grammar rules, known as Universal Grammar (UG), which are shared by all human languages and which all languages have in common. UG suggests that all human languages have the same set of categories and relations. To communicate by language, speakers make up infinite sentences by the use of finite components. From this perspective, a grammar must contain a finite system of rules that generates infinitely surface and deep structures. It must also contain rules that relate these abstract structures to certain representations of sound and meaning. UG can be regarded as innate special structure or property, without which human linguistic system cannot develop. As Chomsky has observed, “Universal grammar is not a grammar, but rather a theory of grammars, a kind of metatheory or schematism for grammar” (Chomsky, 1980). Chomsky treats UG as the initial state of the human brain, as a mental organ. The process of language acquisition could be seen as the development of language faculty under the stimulus of external environments. The initial state(S0) of language faculty provides the basic framework for the acquisition of any human language, as called Universal Grammar. Under certain linguistic environment, the language has undergone several developing stages (such as S1, S2…Sn) and finally reached a stable state (Ss). Compared with other traditional linguistic theory, Chomsky’s UG theory has shifted the focus of linguistic study from linguistic behavior to the internal mechanism of language. UG hypothesis is the very foundation of Chomsky’s theoretical system. The core idea of UG could be summarized as ⅰ. There is an innate faculty of language in the human cognitive system shared by all human beings. ⅱ. The initial state of faculty is the internal factor for human beings to learn a language. UG aims to explain the grammatical rules of deep structure. Particular Grammar is to describe linguistic phenomena, while Universal Grammar discusses how to generate these grammatical rules. ⅲ. UG is shared by all human languages. It is like an abstract software, determining the organizational principle of human languages and constraining what human languages are likely to be or not likely to be. ⅳ. UG focuses on linguistic competence rather than linguistic performance. ⅱ. Universal Grammar and Particular Grammar Chomsky points out that the universals of natural language are reflected by two perspectives: substantive universal and formal universal. Human knowledge consists of two parts as well: one is shared by all human languages, known as Universal Grammar; the other is specific for different nations, called as Particular Grammar (PG). UG is acquired by biological evolution and genetic inheritance, while PG is learned under certain linguistic environment. The initial state of language mechanism is equal for all human beings, while it changes with the change of stimulus mechanism and environment and finally reaches stable states which are distinctive for individuals. UG is the theory to investigate initial state, and PG is concerned with the stable state. On the relationship between UG and PG, Chomsky holds that PG is derived from UG by being ‘selected’ or ‘triggered’ after ‘nurturing experience’. In other words, UG exists beyond PG or even without PG. Human beings being capable of particular languages is mainly attributed to the UG. Whatever their nations, race, sex or skin color, or whatever kind of language they acquire, they share the same UG when they were born. Without UG, human beings cannot master a particular language merely on the basis of language materials. UG is the foundation of PG. Grammars of an individual language are all derived from unique UG. UG would transfer to PG under the trigger of specific language materials or language environment. Only when UG transfers into PG, have individuals gained particular language. In other words, whether a person is capable of one or more languages depends on the triggers he received from language materials. Like other biological organs, linguistic competence would flourish with age and reach a mature period. So, UG is the determinant factor for human beings to master the language, with language material and environment a stimulant function. Ⅲ. Reflections on the application of UG on SLA In order to explain how language faculty affects children’s acquisition of their mother tongue, Chomsky proposed a language acquisition model based on UG. In the model, children’s acquisition process of mother tongue has been described as a system composed of principles and parameters (Chomsky, 1981). Principles are innate in the human brain; parameters reflect the distinction of different languages. In the children’s language acquisition, parameters need to be triggered by positive evidence to set the value of parameters for different languages. Take pro-drop parameter as an example. English is not a pro-drop language, which means neither subject nor object pronouns can be silent. (1) a. He speaks English. English b. Parla. (speak.) Italian c. 吃过了。（have eaten） Chinese In the English language, a sentence is ungrammatical without the subject ‘he’, but in the Italian and Chinese language, it is natural to produce sentences without a subject in b and c. Thus, both Italian and Chinese belong to the pro-drop language. Pro-drop is one of the parameters that indicate the difference between languages. Chomsky formulated the relationship between UG and PG in terms of parameters: PG=α· UG. (UG: Universal Grammar, PG: Particular Grammar, and α: parameter) UG is the intrinsic linguistic knowledge, while PG results from the setting or selection of the parameters. Initiating from UG, children’s language acquisition system sets up and defines the parameters of the interlanguage through their learning experience and finally construct PG. Though UG is aimed to explain the acquisition of the mother tongue, it provides a significant theoretical framework to study the linguistic structure and acquisition mode of the human brain. Under the framework, it becomes possible to compare the acquisition process of different languages, so more and more researchers put the spotlight on the application of UG approach on the interpretation of second language acquisition (SLA). Scholars are particular concerned whether UG could also work on SLA, in other words, whether the second language (L2) learners have access to UG. Generally, there are three views in regards to the accessibility of UG to L2 learners: 1) No access: Clahsen & Muysken (1986) and Meisel (1991) hold that UG principle and parameter could not help in SLA. SLA is different from the acquisition of children’s mother tongue. Children acquire their mother tongue by innate linguistic competence or Faculty of Language, while adults’ second language learning depends on general learning strategies. The acquisition of the mother tongue belongs to the field of linguistics, while SLA the field of Cognition; 2) partial access: Schachter (1988) and Clahsen & Muysken (1989) believe that on the one hand, L2 learners have access to UG which all languages have in common. Under the constraints of the UG, adult learners would not make wild grammatical mistakes. On the other hand, parameters cannot be acquired completely. Adults can only acquire those same parameters as mother tongue. In other words, it is safe to say adult partially acquire the competence of target language. 3) full access: In the view of Flynn & Manuel (1991) and Schwartz & Sprouse (1996), L2 learners own the same linguistic competence as native speakers, which means SLA have full access to the UG. Full access view argues that linguistic knowledge of L2 learners far outpaces the input they receive, which means the logical problems in the acquisition of mother tongue will also occur in SLA. What’s more, the linguistic knowledge of L2 learners is structure-dependent. It is the hierarchical structures of the sentences that linguistics relies on rather than linear word cluster. To show the accessibility of UG to SLA, White & Juffs (1998) implemented an empirical study on the acquisition of wh-island conditions constraints. They based their experiment on the condition that if Language 1(L1) is opposite to Language 2(L2) on a parameter of grammatical principles, the grammar of L1 could not provide any information of parameter for the L2 learners. In other words, the Chinese language has an opposite parameter setting to the English language on the island constraints. English is a language which preposes their wh-phrases, while in Chinese, wh-phrases stay in situ. If results indicate that English language learners whose mother tongue is Chinese could abide by wh-island condition, it could be safe to conclude that L2 learners derive the knowledge of the parameter from the continuing function of UG constraints in interlanguage grammar rather than L1, Chinese language. The results of the experiment showed that all L2 learners obeyed island condition constraints during English language learning, indicating that the parameters of UG have been reset during the SLA. The study verified the constraint function of UG to SLA. UG works on both mother tongue and L2. White (2007) clarified misunderstanding on the SLA study in the framework of generative grammar in terms of the research scope, mother tongue transfer and methodology. Based on the previous study on the application of UG to SLA, my reflections are as follows: First, the research scope of SLA is neither confined to the description of language use nor the explanation of observed linguistic phenomena. SLA research is aimed to investigate the linguistic competence or knowledge indications of interlanguage to reveal the nature of the human mind. It is not scientific to search for the process of SLA on the observation of some empirical study. Second, adults’ failure in SLA is a rather complicated question which cannot be attributed to a single factor, such as UG. The most common misunderstanding is that if UG governs interlanguage grammar, there would be no difference between the final stable state of L2 and L1. Nevertheless, in reality, most L2 learners have been keeping in developing state of interlanguage and far away from the stable state. Such factors as the construction of the L1 knowledge system, the input quality of second language and emotions of the L2 learners may contribute to the failure of SLA. Besides, SLA could not be seen as one aspect of grammar acquisition, for it is easier to identify phonetic, pragmatic and lexical phenomena than grammar. It is somewhat farfetched to determine the success or failure of SLA merely based on UG. Third, we find no way to make a clear, unified and widely-accepted explanation for UG, as difficult as to define what language or grammar is. UG is an abstract system of principles and parameters for deduction and derivation but not for generalization and summarization. UG is still developing, with the study of SLA based on UG approach an experimental attempt. Realizing the tentative and developing nature of UG, we can have a more correct and overall understanding of SLA. Fourth, as an experimental testing instrument, grammaticality judgement employed by SLA research arouses a great dispute. Though grammaticality judgement can reflect the linguistic competence of the participants, if designed inappropriately, participants may be overly aware of testing purposes or adopt their grammatical knowledge consciously. Of course, it is a problem facing all SLA researches other than UG approach. Ⅳ. Confusions in UG ⅰ. Methodology Lin (2015) argued that the fatal problem with UG lies in the methodology of UG by showing the method used by UG theorists in obtaining subjacency. UG grammar attributes the constraints on the movement of words in sentences to human brain structure. However, other causal factors, such as memory, attention, processing speed, are likely to determine the constraints as well. It is not scientific to adopt the method of finding subjacency merely on the basis of grammatical and ungrammatical sentences, but without taking those causal factors into consideration. UG grammar hypothesizes a variety of innate language universals, which include both syntactic principles and parameters, which are obtained with the same unscientific method as that adopted in formulating subjacency. UG could not be viewed as a scientific approach due to the lack of consideration or neglect of causal factors in research methodology. No real innate language universals can be found using this method. (Lin, 2015) Besides, Lin (2019) argued that UG also suffers from ‘theory-internal’ problems. In UG, there is no model to justify whether a law has been found within UG. In this sense, it is impossible to find any universal, innate, human language principles. The methodology adopted by UG on the basis of the syntactical analysis is not as scientific as other natural science and somewhat misleading. ⅱ. UG and Rationalism UG prioritizes innate linguistic competence of initial stage, which is the common ground and source of different human languages. Linguistic competence of stable stage, which is decided by the former, is the secondary. Even though UG is considered as a revolution in the linguistic field of the twentieth century, a similar linguistic view emerged as early as the Middle Ages. Greek and Latin philosophers put forward the ‘innateness’ view, which influenced the Rationalism of France and Britain in the eighteenth century. However, scholars held a skeptical attitude towards the possibility of universal grammar for the language structures other than Latin-based languages and finally abandoned the ‘innateness’ view. Until the twentieth century, Chomsky’s TG grammar and UG aroused the interest of people and were given new defined terms. UG theory did not separate itself from Rationalism idea. (Yang, 2004) Rationalists initiated their methodology from ‘innateness’ view. They believed in the priority of innateness and reason. In Chomsky’s view, UG is an innate notion shared by all human beings, which could be specifically applied to different languages. Yang (2004) stated that the UG theory is in nature the revival of Rationalism in Philosophy. It is difficult to identify the misconception of UG for we tended to separate linguistics from philosophy and thus failed to examine UG from a philosophical view. What Chomsky wants to interpret by UG is still an old philosophical question-where did human knowledge and competence come from and whether it is innate or cultivated. ⅲ. Logical Problems Here I would like to present some logical problems that are confusing: First, Chomsky argued that children would experience several stages such as the babbling stage, nonsense word stage, holophrastic stage, but how could he claim at the same time that children can acquire their mother tongue very fast and with little effort? Besides, in most cases, it will take at least five or six years for children to completely acquire a language, which actually cannot be regarded as a short time. Second, language is a kind of competence. If human beings are born with linguistic knowledge and competence, what about the competence of other brain organs, such as logical and imaginary competence? Besides, children also bear the ability to distinguish color, to paint, to dance, and to sing when they were born. Is it possible to infer that human beings are born with universal dancing competence or universal singing competence? Third, one of the grounds for UG is that different human languages share common features, which results from the basic cognitive ability and physical community space shared by all human beings. Is it convincing to base the innate UG on those features? What are the functions of practice and experience in human cognitive activities? Fourth, if we remove some local features of a language, for example, article, interrogative word or relative pronoun in the English language, but retain only lexical forms, such as nouns, verbs or adjectives, English could be logically formalized without English features and contents, leaving only a pure form structure. Grammar has been transferred to logics which could be eradicated, then does UG still exist? ⅳ. Limitations of UG Since Chomsky proposed the innateness hypothesis in 1957, it was in contention in the linguistic world. Instead of being supporters or opponents, more held a skeptical attitude toward his theory. Is there a language faculty in the human brain? In which part of the brain does it lie? Are children born with Universal Grammar? How does UG help children acquire language? For those questions, neither did Chomsky justify them by providing physiological and anatomical proof, nor gave a definite description of how language faculty work. Without answering those questions, the reliability of UG theory cannot be convincing. Another limitation of the UG theory lies in its over-emphasis on the innateness of language acquisition, and the linguistic environment is only treated as a parameter with triggering function. It is a conscious ignorance of the linguistic environment in language acquisition. In some sense, while Chomsky refuted the language acquisition approach of Behaviorism, he exaggerated the innate factors of language acquisition, another extreme. Investigating the universal grammar of human language based on language form is another problem in Chomsky’s theory. Chomsky focuses on the internalized language(I-language) and attempts to find out abstract regularities of language form by surveying a particular language. However, it is insufficient to discuss language universals or even explore the human mind on the basis of syntax study of a particular language. As a result, Chomsky would not achieve much on the universal grammar of all human languages with the study of linguistic form. Chomsky’s theory has undergone development and evolution, but universals of language remain unchanged as his ultimate research purpose. Chomsky believes that UG has the maximum interpretative force to uncover the design features of all human languages, which can be observed from different stages of his theory. From Generative Grammar to Universal Grammar, he used ‘parameters’ to replace rules and then devoted himself to the Minimalist Program (MP). Due to the striking difference between world language, it is difficult or even impossible to explain all languages by UG. UG leaves much to be desired in justifying those questions. Ⅴ. Conclusion UG is a distinct and representative theory based on hypothesis, which brings about supports as well as attacks. But without any doubt, it is of great value to the modern linguistic study. Since its birth, UG has never ceased to develop and explore human language, which shows its strong vitality. When Chomsky denies himself during the research, he aims to achieve more and better. UG is highly generalized, clarified and generative. Chomsky holds that UG exists in the human brain, which indirectly and positively contributes to the study of the human brain and artificial intelligence. What’s more, the study on the application of UG to the acquisition of L1 and L2 has been in progress, with new foreign language teaching methods or materials emerging on the basis of UG, such as cognitive psychology, communicative approach to teaching L2, error analysis, etc. Whatever UG is, the attempt to investigate the universal grammar of language indicates that human beings intend to explore the universal rule of intelligence structure. References Chomsky, Noam. 1965. Aspects of the theory of syntax. MA: MIT Press, Cambridge. Chomsky, Noam. 1980. Rules and Representations. Columbia University Press, New York. Chomsky, Noam. 1981a. Lectures on Government and Binding. Foris, Dordrecht. Chomsky, Noam. 1981b. Principles and parameters in syntactic theory. Hornstein, N., Lightfoot, D (eds.). Explanation in linguistics: The logical problem of language acquisition, Longman, London. Clahsen, H&P. Muysken. 1986. The availability of Universal Grammar to adult and child learners-the study of the acquisition of German word order. Second Language Research. Clahsen, H&P. Muysken. 1989. The UG paradox in L2 acquisition. Second Language Research. Flynn, S & S. Manuel. 1991. Age-dependent effects in language acquisition: An evaluation of the “critical period” hypothesis. L. Eubank. Point Counterpoint: Universal Grammar in the second Language. John Benjamins, Amsterdam. Francis Y. Lin. 2015. What is really wrong with universal grammar (commentary on Behme). Linguistic Society of America. Francis Y. Lin. 2017. A refutation of universal grammar. Lingua Franca. Francis Y. Lin. 2019. Methodological confusions in universal grammar: Reply to Kim. Lingua Franca. Meisel, J. 1991. Principles of Universal Grammar and strategies of language learning: Some similarities and differences between first and second language acquisition, L. Eubank. Point Counterpoint: Universal Grammar in the second Language. John Benjamins, Amsterdam. Schachter, J. 1988. Second Language acquisition and its relationship to Universal Grammar. Applied Linguistics. Schwartz, B & R. Sprouse. 1996. L2 cognitive states and the full transfer/full access model. Second Language Research. White, L. & A. Juffs. 1998. Constraints on wh-movement in two different contexts of non-native language acquisition: Competence and processing. In S. Flynn. G. Marohardjono, & W. O’ Neil(eds.). The Generative Study of Second Language Acquisition. NJ: Lawrence Erlbaum Associates, Mahwah. White, L. 2007. Linguistic theory, Universal Grammar, and Second Language Acquisition in B. Vanpatten & J. Williams(eds.). Theories in Second Language Acquisition. NJ: Lawrence Erlbaum Associates, Mahwah. Xiuzhen Yang. 2004. Does Universal Grammar really exist? Foreign Language Teaching Association. Zhi Li & Manchun Dai. 2009. Impact of Chomsky’s UG Theory on Second Language Acquisition Research. Shandong Foreign Language Teaching Journal. Zhuanglin Hu, 2006. Linguistics A Course Book. Beijing University Press, Beijing.